Monocular reconstruction of wind-driven vegetation is severely underconstrained: motion along the viewing direction is largely unobservable, a moving canopy offers few reliable correspondences, and nearly the entire scene is dynamic, providing little static reference. Directly-learned deformation fields in 4D Gaussian Splatting therefore optimize photometric consistency rather than recover the motion that produced it. We replace that field with a physically parameterized deformation prior: one damped harmonic oscillator per rigid part, driven by the observed wind and integrated by differentiable RK4, supervised photometrically alone. To test whether such a prior is physically grounded rather than merely well fit, we build a controlled synthetic testbed of three procedurally generated trees spanning an order of magnitude in skeleton complexity, whose per-part natural frequency follows from its own geometry and whose damping ratio is a fixed constant, both held out of training. On it, we measure held-out views, temporal extrapolation, zero-shot transfer to unseen wind speeds, and recovery of the physical parameters themselves. The prior costs appearance fidelity on in-distribution views and extrapolates markedly better outside the training window and the training wind, while parameter recovery is far weaker than it first appears: frequency recovery survives an untrained null control on only the sparsest of the three trees, and damping is not recovered at all.
Phenotyping an agricultural crop is crucial for studying its entire life cycle, as it provides vital insights to improve yield and, ultimately, food production. Doing the same for crops grown on remote sites is a challenge for the specialists who cannot be available on-site. 3D reconstruction techniques offer a promising solution to this problem by enabling crop digitization, allowing specialists to access the resulting 3D crop models from anywhere at any time. In this work, we evaluate recent 3D reconstruction pipelines for crop phenotyping. We focus on 7 mesh reconstruction pipelines and measure the fidelity and consistency of their outputs qualitatively and quantitatively. Our results suggest that the meshes produced by the GGGS, PGSR, and 2DGS are preferable to the other pipelines, owing to their quantitative metrics and visually pleasing outputs. The GGGS pipeline is better than the second-best pipeline (2DGS) by about 27\% on the radar chart with 5 dimensions, namely, User ratings, Chamfer distance, LPIPS, PSNR, and SSIM.
Site-specific spraying in chili pepper production requires organ-level segmentation of leaves, peppers, and flowers from handheld field images. However, RGB appearance becomes unreliable under organ overlap, occlusion, dust, and fruit specularity. Offline monocular depth from Depth Anything V2 provides an accessible structural prior without RGB-D sensing, but its reliability is spatially concentrated rather than uniform. To address this, we propose Depth-Routed Selective Attention (DRSA), an estimated-depth-guided segmentation network. DRSA predicts a single per-pixel routing field that, through one shared decision, jointly governs where depth-boundary cross-attention and residual depth fusion contribute, so geometric cues act near organ contours while RGB remains the default carrier. The routing field is calibrated online from depth-on and depth-suppressed predictions without trust-map annotation. We construct PepperField-EstDepth, a self-built dataset of 3, 940 handheld field images paired with estimated monocular depth, on which DRSA achieves \(90.20\%\) mIoU and \(84.48\%\) boundary mIoU, outperforming both RGB-only baselines and attention-based RGB-D fusion baselines; over the RGB segmentation reference, the gains are \(+1.98\) and \(+2.67\) percentage points, respectively. Under group cross-validation, DRSA reaches \(0.8919\pm 0.0031\) mIoU and \(0.8294\pm 0.0046\) boundary mIoU. For the spraying application, DRSA attains a target recall of 0.9814, a target precision of 0.9756, and an organ-level off-target activation of \(2.44\%\) . Visual-pipeline timing shows a segmentation-only latency of 43.0 ms under pre-generated estimated depth, rising to 219.6 ms for the RGB-to-mask visual pipeline when online Depth Anything V2-L depth generation is included. These results support DRSA as a pre-spray organ-level perception module.
Herearii Metuarea · Abdoul-Djalil Ousseini-Hamza · Walter Guerra · Francesca Zuffa · Francesco Panzeri · Andrea Patocchi · Lidia Lozano · Shauny Van Hoye · François Laurens · Jeremy Labrosse · Pejman Rasti · David Rousseau
Abstract In machine learning-driven plant phenotyping, well-annotated image datasets are essential for developing robust models capable of capturing phenological variability across environments. Here, we introduce DeepPhenoTree-Apple Edition , a multi-site and multi-variety RGB image dataset dedicated to the detection of key phenological stages in apple trees. The dataset includes 48,320 time-stamped RGB images acquired across four European orchards of the Apple REFPOP consortium under biological, agronomic, and environmental variability, including differences in genotypes, orchard architectures, phenological development, temperature, and humidity conditions. From this large corpus, a carefully curated subset of 808 representative images was manually annotated. It includes 241,600 expert annotations covering developmental stages from dormant bud to fruit maturity. Images were acquired using a standardized tractor-mounted phenotyping platform equipped with active flash illumination. Active flash illumination was used to reduce illumination variability and homogenize exposure, shadows, and sunlight differences across sites. Phenological structures were annotated following the BBCH scale, with bounding boxes adapted to organ visibility and developmental stage. In addition to the dataset, we provide deep-learning baseline experiments to illustrate detection performance and detection performance across locations.
Early and real-time detection of rice leaf diseases (RLD) poses a significant challenge for farmers, especially in rural regions with limited access to advanced technology. Conventional deep learning models often require substantial computational resources, rendering them impractical for deployment on mobile or edge devices commonly used in agricultural environments. Although models such as ResNet50, VGG16, and InceptionV3 can achieve high accuracy, they are computationally expensive and may be less suitable for real-time deployment in smart agricultural systems. Furthermore, many existing studies rely on limited datasets, which can restrict model generalizability under diverse real-world conditions. Although transfer learning can improve classification performance, developing lightweight models suitable for real-time IoT deployment remains challenging. To address these limitations, we curated a hybrid dataset of 7,092 rice leaf images by combining self-collected and Kaggle samples and proposed IoT-RiceMobileNet, a lightweight improved MobileNetV2-based model that can be effectively integrated into our developed IoT system. The proposed model outperformed transfer learning and deep learning baseline models, achieving 99.19% test accuracy and 99.20% precision while maintaining low computational complexity. We further confirmed model stability using stratified 5-fold and 10-fold cross-validation on both the constructed RLD and multi-source datasets, achieving mean accuracies of 98.04% and 98.13% on the constructed RLD dataset and 98.11% and 98.58% on the multi-source dataset, respectively. Moreover, our models compact size and 85.17 FPS inference speed support real-time deployment. Finally, the model was integrated into an IoT-enabled mobile, web, and cloud-based inference framework, demonstrating its practical potential for scalable rice disease detection in smart agriculture.
Reproduction assets foundThe paper's self-collected rice leaf image dataset and source code are publicly archived on Zenodo and GitHub, and the third-party Kaggle/Mendeley image datasets used to construct the RLD and multi-source datasets are publicly available. All are paper-specific, public, and actionable.Dataset · publicrice disease monitoring and decision support in precision agriculture.
Supporting information
S1 Appendix
Cross-validation results for the multisource dataset.
(PDF)
Data Availability
The self-collected rice leaf image dataset used in this study is publicly available through the archived IoT-RiceMobileNet repository on Zenodo: https://doi.org/10.5281/zenodo.21140529 . The public third-party datasets used in this study are available from their original sources: https://www.kaggle.com/datasets/dedeikhsandwisaputra/rice-leafs-disease-dataset/data
https://www.kaggle.com/datasets/anshulm257/rice-disease-dataset
https://data.mendeley.com/datasets/hx6f852hw4/2 Third-party images are not redistribOpen asset ↗Zenodo · 10.5281/zenodo.21140529lines:868-883Code · public://www.kaggle.com/datasets/anshulm257/rice-disease-dataset
https://data.mendeley.com/datasets/hx6f852hw4/2 Third-party images are not redistributed in the Zenodo archive and should be obtained from their original sources according to their respective licenses. The source code is publicly available through the GitHub repository: https://github.com/eimran/IoT-RiceMobileNet and is also archived on Zenodo at: https://doi.org/10.5281/zenodo.21140529 .
Funding Statement
The author(s) received no specific funding for this work.
References
1. Sokra I, Somaly S, Meta H, Sarun H, Molikoy C. Factors affecting rice production: A systematic review. J Agric Technol. 2026;2(1):19–46. doi: 10.6084/m9.figshaOpen asset ↗GitHub · eimran/IoT-RiceMobileNetlines:868-883Dataset · publicthe multisource dataset.
(PDF)
Data Availability
The self-collected rice leaf image dataset used in this study is publicly available through the archived IoT-RiceMobileNet repository on Zenodo: https://doi.org/10.5281/zenodo.21140529 . The public third-party datasets used in this study are available from their original sources: https://www.kaggle.com/datasets/dedeikhsandwisaputra/rice-leafs-disease-dataset/data
https://www.kaggle.com/datasets/anshulm257/rice-disease-dataset
https://data.mendeley.com/datasets/hx6f852hw4/2 Third-party images are not redistributed in the Zenodo archive and should be obtained from their original sources according to their respective licenses. The source code is Open asset ↗Kaggle · dedeikhsandwisaputra/rice-leafs-disease-datasetlines:868-883Dataset · publicataset used in this study is publicly available through the archived IoT-RiceMobileNet repository on Zenodo: https://doi.org/10.5281/zenodo.21140529 . The public third-party datasets used in this study are available from their original sources: https://www.kaggle.com/datasets/dedeikhsandwisaputra/rice-leafs-disease-dataset/data
https://www.kaggle.com/datasets/anshulm257/rice-disease-dataset
https://data.mendeley.com/datasets/hx6f852hw4/2 Third-party images are not redistributed in the Zenodo archive and should be obtained from their original sources according to their respective licenses. The source code is publicly available through the GitHub repository: https://github.com/eimran/IoT-RiceMOpen asset ↗Kaggle · anshulm257/rice-disease-datasetlines:868-883Dataset · publicived IoT-RiceMobileNet repository on Zenodo: https://doi.org/10.5281/zenodo.21140529 . The public third-party datasets used in this study are available from their original sources: https://www.kaggle.com/datasets/dedeikhsandwisaputra/rice-leafs-disease-dataset/data
https://www.kaggle.com/datasets/anshulm257/rice-disease-dataset
https://data.mendeley.com/datasets/hx6f852hw4/2 Third-party images are not redistributed in the Zenodo archive and should be obtained from their original sources according to their respective licenses. The source code is publicly available through the GitHub repository: https://github.com/eimran/IoT-RiceMobileNet and is also archived on Zenodo at: https://doi.org/10.5Open asset ↗hx6f852hw4/2lines:868-883Code / dataset availability confirmedOpenAlex · Crossref · checked 17 Sept 2026
Junjie Wu · Bei Zhou · Jie Liu · Zhuyang Xie · Jie Luo · Jiajun Liu · jiahui zhu · Mingju Li · Yan Ai · Mingwei Liu · Yu Chen
MaizeRapeseed / canolaSoybeanLiDAR / point cloudLeafStem / branch
Three-dimensional plant point clouds retain crop architecture, but organ-level phenotyping requires reliable semantic separation of leaves and stems. Plant-GeoAT is a geometry-aware parsing network that encodes local relative-XYZ neighbourhoods before RGB fusion and couples spatial neighbourhoods with feature–space relations for dense point prediction. We evaluated the model separately within the native protocol of a self-built structure-from-motion rapeseed dataset, an image-based soybean dataset, and laser-scanned Pheno4D maize and tomato datasets; these are within-dataset train/test experiments, not cross-dataset transfer or domain-generalisation tests. Across five seeds, mIoU was 92.26 ± 0.28%, 82.50 ± 0.24%, 99.74 ± 0.05%, and 94.75 ± 0.15%, respectively. Maize Stem IoU reached 99.57 ± 0.09%. Adding LLGE increased the four-dataset average mIoU from 66.38% to 85.84%, and the complete LLGE + SSCA model reached 92.31%. On the fixed six-sample test set, exploratory semantic-guided clustering achieved 86.67 ± 7.45% Count Accuracy; structural correctness ranged from 3/6 to 4/6 samples across seeds. Plant-height consistency was assessed independently on all 60 reconstructed rapeseed samples. The results indicate that geometry-to-context encoding produces organ-level semantic units for subsequent phenotyping, while independent instance-labelled datasets and additional growth stages are still needed for trait-level validation.
Reproduction assets foundThe paper's Plant-GeoAT research-preview code (model implementation, training and evaluation scripts) is publicly available at an immutable GitHub commit under the authors' kimevans111 account. The self-built rapeseed point-cloud dataset and annotations are not public (available only on request), and third-party publicCode · publictheir original providers and publications; this article does not redistribute
third-party datasets. The processed summary metrics required to evaluate the reported results are
included in the main article and Supplementary Material. The reviewed Plant-GeoAT research-
preview code is available at the immutable historical commit https://github.com/kimevans111
/Plant-GeoAT/commit/e0aba5b94e26ba4306d0f7f9901185883fe8d4c4 (accessed on 30 August 2026).
This snapshot is distinct from the repository’s current default branch. It contains the model implemen-
tation, training and evaluation scripts, data-format and release documentation, command templates,
citation metadata, and licence/notiOpen asset ↗Plant-GeoATpdf-raw-page:24 lines:1-50Plant phenotyping relevance matchEurope PMC · checked 17 Sept 2026
Cell division patterns shape developing organs, but their systematic analysis remains limited by the technical difficulty of existing detection methods. Here, we present a practical method for inferring recent daughter-cell pairs from static cell shapes. Although cell divisions have long been inferred by expert judgment, the accuracy of this approach has rarely been quantified. Using live imaging of Arabidopsis leaf primordia, we evaluated 49 features from 7,573 neighboring cell pairs, including 1,114 direct daughter pairs and 6,459 non-daughter pairs. A simple junction-angle criterion correctly inferred 1,030 daughter pairs with 94.1% precision and 92.5% recall, whereas symmetry-based indices performed poorly. Temporal analysis revealed that junction angles were initially high after division, decreased during growth, and were altered by subsequent divisions, thus, separating daughter pairs from non-daughter pairs. This method also achieved 98.4% precision and 97.2% recall in an independent shoot apical meristem live-imaging dataset. Application to the pointed-tip Arabidopsis mutant rpl4d-3 and two non-model species further supported its potential applicability for analyzing division patterns in diverse leaf morphologies. For community use, we provide FIJI/ImageJ and Qt-based graphical user interface implementations. Overall, this study quantitatively validates empirical geometric cues and provides a practical framework for inferring daughter pairs from static tissue images.
Premise Herbarium specimens are increasingly used to extract morphological traits for ecological and evolutionary studies, yet the effects of tissue desiccation on trait measurements remain poorly understood. Here, we tested two hypotheses: (H1) whether higher tissue water content leads to greater measurement changes after herborization and (H2) whether fresh trait values can be reliably predicted from measurements of herborized specimens. Methods We evaluated the reliability of herbarium-based measurements by comparing fresh and dried traits of leaves, flowers, fleshy fruits, and seeds across 262 individuals representing 133 Neotropical Myrtaceae species. Phylogenetic least squares models and machine learning regressions were used to test H1 and H2. Results Leaves and flowers generally shrank after herborization, fruits size metrics tended to increase, and seeds were largely unaffected. Water content was significantly associated with the magnitude of herborization effects in flowers and some leaf and seed traits. Fresh trait values were accurately predicted from measurements of herborized specimens. Prediction errors were lowest for leaf traits, followed by fruits, flowers, and seeds. Discussion These results partially support H1 and support H2, indicating that herborized specimens can be reliably used for trait analyses when organ-specific responses are considered, providing a practical framework to account for potential desiccation bias in functional trait research.
Reproduction assets foundThe paper's trait measurements (Appendix S6 species mean data) and analysis code are openly available in the authors' GitHub repository, explicitly stated in the Data Availability Statement.Code · publice also acknowledge the Sociedade Botânica do Brasil (SBB) for financial support through the Scientia Amabilis 2024 grant.
DATA AVAILABILITY STATEMENT
The data supporting the findings of this study are available in the Supporting Information of this article (Appendix S6 ). Additionally, all data and code are openly accessible at https://github.com/ykilsztajn/fresh_dry_myrtaceae .
REFERENCES
Bewley , J. D.
,
K. J.
Bradford
,
H. W.
Hilhorst
, and
H.
Nonogaki
. 2012 .
Development and maturation
. In
Seeds: Physiology of development, germination and dormancy , 3rd ed. , 27 – 83 .
Springer
,
New York, New York, USA
.
Blonder , B.
,
V.
Buzzard
,
I.
Simova
,
L.
Sloat
,
B.
Boyle
,
R.
Lipson
,
B.
AguiOpen asset ↗ykilsztajn/fresh_dry_myrtaceaelines:268-513Plant phenotyping relevance matchCrossref · checked 17 Sept 2026
Published15 Sept 2026Black Sea Journal of Agriculture
Leaf area (LA) is one of the most important morphological traits for the assessment of plant growth, biomass production and physiological performance. For forage crops, rapid, accurate and non-destructive methods of LA estimation are particularly important to allow repeated measurements throughout the growing season. The aim of this study was to compare the leaf area estimation of sainfoin (Onobrychis viciifolia Scop.) by four regression models (linear, logarithmic, polynomial and multiple linear regression) based on simple plant morphological measurements. Ninety sainfoin plants were evaluated at the flowering stage under controlled conditions. Plant height and canopy width were measured manually and actual leaf area was measured with a LI-COR LI-3100C Leaf Area Meter. Regression analyses were performed using SPSS software to construct prediction models and compare the performance of the models. Of the models tested, the multiple linear regression model had the highest prediction accuracy (R²=0.981), which was significantly better than the polynomial (R²=0.413), linear (R²=0.399) and logarithmic (R²=0.361) models. The results demonstrated that the combination of plant height and canopy width greatly improved the leaf area estimation over the use of a single predictor. The predictive performance of simple regression models was low. Multiple linear regression was a reliable, rapid and non-destructive method to estimate the leaf area of sainfoin. These results pave the way for practical morphological models that could assist plant growth monitoring and agronomic studies while limiting the number of destructive samplings.
Traditional weight measurement methods need cutting or weighing of the fruit, and this is not practical for preharvest evaluation and market transactions. The suggested approach gives a practical solution for farmers and sellers by giving accurate weight predictions using only external characteristics. Modern technologies, especially artificial intelligence, data analytics and machine learning, are making big changes in the agricultural area. One of the machine learning models used in agriculture is Artificial Neural Networks (ANN). ANN became an important tool in analyzing agricultural data, predicting plant growth processes, finding diseases, determining the effects of environmental factors, reaching productivity goals and detecting weeds and harmful plants. A dataset is created by comprehensively examining 52 watermelons. The dataset includes acoustic properties, geometric measurements, and visual characteristics. The dataset is divided into 40 training samples and 12 test samples. Balanced representation is ensured by using stratified sampling when selecting test samples. The Min-Max normalization method is used for data preprocessing. Nine different training algorithms are comprehensively evaluated within the scope of the study. Eight critical parameters of the ANN models are optimized using four different optimization algorithms to increase the accuracy rate and avoid overfitting. Artificial Bee Colony (ABC), Artificial Fish Swarm Algorithm (AFSA), Whale Optimization Algorithm (WOA) and Grey Wolf Optimizer (GWO) are used as optimization methods. Assessed by five-fold cross-validation, the best configuration (One Step Secant with WOA) achieved a mean absolute error of 0.92 ± 0.28 kg and an RMSE of 1.23 ± 0.37 kg, corresponding to about 10% of the mean fruit weight. Developing a real-time mobile application using the optimized best model will provide practicality in large-scale agricultural enterprises, controlled environments such as greenhouses, and agricultural markets.
Mohammadreza Zare · Sina Jamalzadegan · Aditi Dey Poonam · Syed Ahmed Jaseem · Jeong Yong Kim · Jin Xu · Hanbin Choi · Nurit Atar · Baihang Chen · Michael D. Dickey · Qingshan Wei
Abstract Wearable sensors are essential tools for the continuous, real-time monitoring of human and plant health. They interact directly with skin and living plant organs to track physical, electrical, and chemical signals that are often difficult to measure with off-body or off-plant sensors. Sensors that are optically transparent minimize interference with the host. Achieving transparency is challenging because devices must minimize light absorption, haze, and color tint while maintaining good conductivity, mechanical flexibility, strong adhesion, and breathability. This Review focuses on transparent wearable sensors in which optical transparency is functionally important for living-interface operation, including optical access, unobtrusive wear, multimodal readout, or reduced perturbation of plant photosynthesis. We include device-level demonstrations that integrate transparent or semitransparent substrates, electrodes, encapsulants, or sensing layers with human skin, microneedle interfaces, or plant organs, while excluding opaque wearables, nonwearable transparent electronics, and rigid transparent electrodes without sensing integration. We categorize different transparent components needed for a wearable sensor, including transparent insulators, electrodes, and sensing materials, and relate these components to human use and emerging plant health applications. In contrast to prior reviews, which largely treat transparency as a material property and survey human or plant wearables separately, we frame transparency as a system-level design requirement that unifies both domains, clarifying where optical neutrality is functionally necessary and what should be reported to make transparency a comparable design variable.
Proline is an important biomarker in biological regulation and plant stress adaptation, increasing the demand for highly specific analytical detection methods. Herein, we report the use of DMAC as a chemodosimetric probe that exclusively identifies proline in the presence of 19 competing amino acids. The sensing mechanism relies on an irreversible chemical reaction between proline and DMAC, producing a distinct optical response with negligible background interference. This exhibits a highly sensitive limit of detection of 15.8 nM (1.82 ppb), signifying its importance in trace-level quantification. We thoroughly validated the underlying reaction mechanism using a combination of experimental and theoretical techniques. To demonstrate real-world applicability, the DMAC probe was integrated into a microfluidic device, resulting in a portable device designed for rapid, on-site analysis. Finally, the sensor's practical utility was demonstrated by successfully monitoring proline dynamics during plant growth.
Accurate 3D phenotyping of rice in multi-plant scenarios is important for evaluating yield-related traits and supporting high-throughput breeding. In particular, grain-level instance segmentation from point clouds provides a direct basis for quantifying panicle structure and grain number. However, this task remains challenging because densely grown rice plants often exhibit severe inter-plant occlusion, organ adhesion, and highly similar local structures, making it difficult to simultaneously capture global plant organization and fine-grained grain geometry. To address these challenges, we propose SPINet, a structure-enhanced point cloud instance segmentation network for 3D phenotypic extraction in multi-plant rice scenes. The proposed framework introduces a RiceMamba2 encoder that combines the long-range dependency modeling capability of Mamba2 with a parallel convolutional branch for sequence-neighborhood enhancement, enabling effective representation of both global spatial context and local organ-level details. In addition, a center prior-guided decoder is developed to provide explicit spatial anchors for instance queries, thereby improving the separation of adjacent and adhered rice organs in dense scenes. To further enhance training stability and convergence, a DINO-based denoising strategy is incorporated into the optimization process. Experimental results on the controlled indoor Grouped Rice Dataset (GRD) show that SPINet achieves an average precision (AP) of 65.73%, outperforming OneFormer3D by 31.38 percentage points. Furthermore, grain counting based on the predicted instances obtains a coefficient of determination ( R 2 ) of 0.8532, demonstrating the potential of SPINet for downstream yield-related phenotypic trait extraction. These results suggest that SPINet provides a promising framework for automated 3D rice phenotyping in complex multi-plant scenarios.
Nadeem Fareed · Carlos Alberto Silva · Alexander J. Gaskins · Susan J. Prichard · Andrew T. Hudak · Jinyi Xia · Cesar Ivan Alvites Alvites Diaz
Field / plotLiDAR / point cloudLeafStem / branchMorphology / geometry measurementSegmentationArchitecture / morphology / geometry
Terrestrial Light Detection and Ranging (LiDAR) provides detailed three-dimensional (3D) observations of forest structure, yet transferable wood–foliar semantic segmentation remains challenging because of forest structural heterogeneity, occlusion, and variability across forest ecosystems and terrestrial LiDAR platforms. Existing Deep Learning (DL) approaches are commonly developed for localized forest conditions or complex multi-class semantic taxonomies, often limiting their transferability to structurally distinct forests. Here, we introduce points2SBL, a geometry-driven, terrestrial LiDAR platform-agnostic DL framework that reduces heterogeneous forest vegetation to a transferable wood–foliar representation and subsequently uses these semantics as the foundation for 3D forest fuel characterization. The framework was developed using globally harmonized benchmark datasets acquired across multiple terrestrial LiDAR systems and broadleaf, coniferous, mixed, regenerating, and structurally complex forests. Three point-based architectures (PointNet++, PointNeXt, and Point Transformer) were benchmarked under identical training and evaluation protocols. Point Transformer provided the most consistent cross-dataset performance, achieving overall accuracy (OA) of 92–94%, mean Intersection-over-Union (mIoU) of 0.78–0.87, macro F1-score of 0.87–0.93, and Matthews correlation coefficient (MCC) of 0.74–0.86 across five independent benchmark datasets. Species-level, vertical-profile, and qualitative cross-ecosystem assessments further demonstrated consistent preservation of wood–foliar semantics throughout the 3D forest fuel continuum, including previously unseen forest point clouds, while revealing that some apparent semantic disagreements originated from manual annotation of benchmarks. Finally, the points2SBL wood–foliar semantics were geometrically decomposed using a geometric algorithm named “Points2woodyseg” into fuel-related components: stem, branches, leaf, surface wood, and surface foliage. Collectively, these results demonstrate that a simplified wood–foliar abstraction can provide a transferable semantic foundation for characterizing structurally heterogeneous terrestrial LiDAR point clouds and deriving essential 3D forest fuel information across diverse forest ecosystems.
Introduction Accurate extraction of maize seedling phenotypes is essential for early growth assessment, yet field-grown seedlings exhibit complex backgrounds, non-rigid leaf deformation, self-occlusion, and adhered leaves that challenge three-dimensional point-cloud analysis. Methods Here, we developed a workflow integrating PMSPA-Net, an enhanced PointNet++ model, with geometric region decomposition for semantic and organ-level segmentation of field-grown maize seedlings. Multi-view RGB images of 65 complete plants were reconstructed with OpenMVS, and all data were partitioned at the whole-plant level before point-block sampling. Fifteen plants were retained as a fixed independent test set, while the remaining samples were used for training and validation across three predefined random seeds. Results Under a common protocol, PMSPA-Net outperformed PointNet, PointNet++, DGCNN, and RandLA-Net, achieving an mIoU of 0.8286 ± 0.0149, a macro-F1 of 0.8981 ± 0.0238, and a balanced accuracy of 0.9561 ± 0.0223. A four-configuration ablation study showed complementary gains from the multi-scale spatial pyramid attention module and IOA-based hyperparameter optimization. Geometric region decomposition also achieved more accurate and complete adhered-leaf separation than DBSCAN in three annotated cases. Point-cloud-derived leaf length and width agreed well with manual measurements. Discussion This workflow demonstrates the feasibility of integrated semantic and organ-level phenotyping under the evaluated field conditions; validation across additional sites, developmental stages, and acquisition conditions is required before broader generalization.
Diaz-Garcia, L. · Munoz, J. R. · Torres-Lomas, E. · McElrone, A. J. · Sharma, S. · Lupo, Y.
GrapevineRootMorphology / geometry measurementSegmentationGrowth / time-series analysisGrowth / development / phenologyRoot system architectureWater status / transpiration
Root system architecture shapes how grapevine rootstocks take up water and nutrients, yet roots remain the least phenotyped grapevine organ because they are hidden and hard to image. We present a low-cost phenotyping pipeline that pairs custom acrylic rhizotrons (about US$30 each) with a consumer flatbed scanner and BiRefNet, a general-purpose deep-learning model used without training on root images, followed by automated mask cleaning, skeleton-based trait extraction, and soil moisture mapping. We tested it on nine commercial rootstocks scanned 16 times over 42 days after transplanting (DAT), with half under a ten-day water deficit. From 1,108 images we extracted 21 whole-root, depth-resolved, and topological traits. Genotypes differed in nearly every trait and in how they changed over time. Heritability of size and branching traits peaked at 0.92-0.93 between 21 and 31 DAT and fell for width, depth, and convex hull once roots reached the rhizotron walls, defining the best measurement window. The image-derived soil moisture map accurately tracked the deficit and its recovery. Deficit plants shifted new root growth to deeper soil without growing less overall, and the substrate dried fastest around older and denser roots. Root brightness decreased with root age and local moisture, and transport segments (axes serving several tips) were brighter than terminal laterals in every genotype. Root system size was associated with stomatal conductance in well-watered plants, and stomatal recovery after re-watering correlated with new root growth. The pipeline turns simple hardware into a quantitative, time-resolved root phenotyping platform suitable for breeding.
Abstract Leafhopper feeding-damage severity is an important factor in the raw-material grading and final quality of Zijin Chan tea, but conventional assessment relies heavily on subjective visual judgment. This study developed a two-stage workflow combining machine-vision screening with offline near-infrared (NIR) spectroscopic reassessment. First, an improved small You Only Look Once version 8 (YOLOv8s) model localized tea-leaf targets and classified slight, moderate, and severe feeding damage. Physical samples corresponding to machine-vision no-result targets were then reassessed using an NIR model integrating standard normal variate preprocessing, Pearson correlation-based spectral-region selection, the successive projections algorithm, and a support vector machine. On 200 test images containing 4,000 annotated targets, the improved YOLOv8s model achieved precision of 88.6%, recall of 87.7%, and mean average precision at an intersection-over-union threshold of 0.5 of 89.3%, representing improvements of 12.2, 9.3, and 8.7 percentage points over the baseline, respectively. Developed from 240 independent physical samples, the NIR model achieved macro-averaged recall of 94.03% on the cross-validated training set and 95.12% on the spectral model-selection set. When the fixed model was applied to physical samples corresponding to 492 machine-vision no-result targets, 473 targets were classified correctly, yielding a reassessment accuracy of 96.14%. Across all targets, standalone machine vision achieved an end-to-end accuracy of 77.7%, whereas the two-stage workflow achieved 89.5%. These results support the feasibility of combining visual and spectral information for laboratory-based grading. Validation using in situ spectra and samples from multiple batches and seasons remains necessary before online deployment.
Image-based crop phenotyping benefits from image series that capture plant development together with organ-level labels. Such paired data are limited because organ annotation is expensive, and following the same plants over time requires repeated, registered imaging. Existing generative models for plants either synthesize temporal imagery without structural labels or generate labeled images without a temporal dimension. We introduce OneGrow, a latent flow-matching model that jointly models wheat images and their organ-segmentation masks over time. Images and masks share a single frozen image autoencoder. A reveal specifies which content is observed context, so the same model covers tasks such as mask-to-image synthesis, image-to-mask segmentation, and temporal forecasting. For sequences longer than the training window, a sliding-window roll-out generates each new image from the preceding frames, keeping long sequences temporally consistent. We train jointly on a large single-frame wheat dataset and a multi-year temporal dataset, using pseudo-labels from a pretrained segmentation model. We evaluate segmentation and image quality against held-out references and assess the multi-task and temporal behavior qualitatively.
Automated wheat-seed detection and counting play important roles in high-throughput plant phenotyping, seed characterization, and agricultural research. Conventional manual counting is labor-intensive, time-consuming, and susceptible to human error when processing large numbers of seed samples. Recent advances in deep learning and object detection provide opportunities to automate these tasks using conventional RGB images. This study focused on a deep learning-based framework for wheat-seed detection and detection-based counting using a custom red-green-blue (RGB) image dataset. A dataset comprising 832 RGB images containing 3,649 manually annotated wheat-seed instances was developed, with images representing one to ten detached wheat seeds per image. All seed instances were manually annotated using bounding boxes and formulated as a single-class object detection problem. Two lightweight object detection models, YOLOv8n and YOLO11n, were trained and evaluated under identical experimental conditions. The model performance was assessed using precision, recall, mean Average Precision at an Intersection over Union (IoU) threshold of 0.5 (mAP@0.5), mean Average Precision averaged across IoU thresholds from 0.5 to 0.95 (mAP@0.5:0.95), training loss curves, confidence-based performance curves, confusion matrices, and qualitative detection outputs. Both models achieved excellent detection performance on the custom wheat seed dataset. YOLOv8n achieved a precision of 0.9947, recall of 0.9963, mAP@0.5 of 0.9940, and mAP@0.5:0.95 of 0.5284. YOLO11n produced slightly higher performance, achieving a precision of 0.9988, recall of 0.9984, mAP@0.5 of 0.9950, and mAP@0.5:0.95 of 0.5385. Overall, the performance of the two models was similar at mAP@0.5, but at mAP@0.5:0.95 stricter criterion, YOLO11n consistently demonstrated the strongest overall performance. The findings show that lightweight YOLO models combined with RGB imaging provide an effective and easy way for automated localization and counting of wheat seeds. The framework provides a manually annotated RGB wheat-seed dataset and a reproducible benchmark to compare lightweight YOLO models for detection. This study provides a practical foundation for future research on automated seed phenotyping, with future work focusing on external validation using more diverse datasets, multi-class seed-quality assessment, and quantitative evaluation of counting performance.
Diseases that affect plants considerably lower the production of crops and pose a significant threat to global food security, particularly in high-value crops like bell pepper (Capsicum annuum L.). The timely and sensitive identification of the disease is essential in reducing losses related to yield and paving the way to sustainable management of crops. Although automated plant disease diagnosis has been improved by artificial intelligence (AI)-driven image analysis, single-model methods are not always robust or generalisable. The study proposes a hybrid convolutional neural network-random forest (CNN-RF) framework for the early detection of bell pepper leaf diseases. Deep visual features of leaf images were extracted and subsequently used with a pretrained ResNet50 model and then used to classify the images with the help of a RF ensemble algorithm. Experiments were conducted on a publicly available PlantVillage bell pepper dataset comprising 495 leaf images (199 bacterial spot and 296 healthy samples). The classification accuracy of the proposed hybrid model was 97.98 % with high class-wise precision, recall and F1-scores. Receiver operating characteristic (ROC) analysis produced an area under the curve (AUC) of 0.98, indicating excellent discriminative performance. The dataset was evaluated using a stratified 80 : 20 train-test split. The results show that deep feature extraction combined with an ensemble learning method can improve classification accuracy and offer a scalable solution to assist in precisionagriculture and timely disease control. However, the framework was evaluated only on the PlantVillage dataset and has not yet been validated under real-field agricultural conditions. The framework demonstrates potential for extension to multi-disease classification tasks, although further validation on field-acquired datasets is required.
Double fertilization in angiosperms is completed by the fusion of two sperm cells with the egg and central cells. While sperm membrane proteins such as GEX2, DMP8/9, and GCS1/HAP2 have been identified as key regulators of gamete attachment and fusion, their specific ultrastructural roles within the embryo sac have remained unclear owing to the technical limitations of live-cell imaging based on fluorescent markers. Here, we optimized high-pressure freezing and freeze-substitution protocols for Arabidopsis pistils and performed three-dimensional reconstructions using scanning transmission electron microscopy (STEM) and array tomography. We identified an electron-dense, partially thickened structure at the apical region of the egg cell before pollination. This structure disappeared following wild-type fertilization, suggesting it serves as a specialized "docking site" to accommodate pollen tube contents and facilitate subsequent gamete interactions. Comparative analysis of fertilization mutants revealed distinct functional stages in the gamete fusion process. In gex2 mutants, sperm cells failed to establish robust connections when positioned at the lateral side of the egg cell, confirming the requirement of GEX2 for initial attachment. In contrast, dmp8 dmp9 double mutant sperm cells facing each other were observed, exhibiting a more distant position from the egg surface, suggesting that DMP8/9 is involved in maintaining stable membrane adhesion before fusion. Furthermore, although gcs1 sperm cells maintained intimate contact and displayed membrane thinning, they failed to undergo final fusion. Our findings demonstrate that STEM and array tomography are powerful tools for visualizing gamete dynamics, providing an integrated model of the transition from stable adhesion to membrane fusion.
Christopher Mataczynski · Maike Glaser · Stuart Huntley · Penelope I. Higgs
Environmental bacteria have abundant signaling systems wired into complex gene regulatory networks to adapt to fluctuating conditions. In Myxococcus xanthus, starvation triggers a developmental program (specialized biofilm) that produces spore-filled multicellular fruiting bodies surrounded by a distinct quiescent state termed peripheral rods. Fruiting body structure as well as the proportion of cells following each fate can be tuned by a large repertoire of signaling proteins, including numerous orphan histidine kinases. Here, we focus on the histidine kinase TodK, which was previously demonstrated to influence developmental progression. We find that loss of TodK produces distinct developmental phenotypes that vary with environmental conditions. To quantify these effects, we developed an image-analysis pipeline that measures aggregation and fruiting body patterning during development on nutrient-limited agar. These analyses revealed the todK mutant prematurely aggregates particularly at the peripheries of the colony. Under submerged-culture conditions, initial production of aggregates was observed with wild-type timing, but these aggregates displayed accelerated transition to mature fruiting bodies. Overexpression of active TodK completely blocked fruiting body formation. Molecular analyses demonstrated that TodK overproduction suppressed expression of core developmental regulators including FruA and CsgA. Interestingly, protein accumulation of MrpC, a transcription factor necessary for expression of both FruA and CsgA, was not significantly perturbed. These data suggest TodK silences MrpC's ability to activate transcription of key developmental targets. Together, these findings establish TodK as a modulator of developmental progression and demonstrate how quantitative phenotyping approaches can reveal biologically meaningful functions for orphan histidine kinases whose mutant phenotypes might otherwise appear subtle.
PotatoField / plotPhysiological trait estimationGrowth / time-series analysisBiomass / plant weightGrowth / development / phenologyWater status / transpiration
Real-time measurement of belowground tuber growth has not been conducted in field crops. Here, a strain-gauge sensor was used to monitor the growth of a single potato tuber and estimate its daily water loss. Over the 11 days preceding harvest, tuber thickness increased by 0.80 mm, corresponding to a daily gain of 1.4 g, or a 4.2% increase. The greatest diurnal fluctuation was 0.439 mm, corresponding to a transpirational water loss of 9.1 mL, or 2.9% of the tuber’s water content. Daily estimated transpiration showed a positive correlation with estimated vapor pressure deficit. This sensor may enable more precise input control and higher temporal resolution than current methods for below-ground crops, supporting improved crop management, yield prediction, and harvest decisions.
This study aimed to evaluate the growth responses of kenaf (Hibiscus cannabinus L.) under nitrogen and compost treatments and to develop a UAV-based plant height estimation model. Ground-measured plant height differences were not significant at harvest (110 days after sowing, DAS), highlighting the need for multitemporal monitoring. Multispectral drone imagery was acquired at five growth stages (20–110 DAS). Object-based image segmentation was applied to extract pure vegetation areas, and digital surface model differencing (ΔDSMt) was used to reduce micro-topographic effects. A UAV-based multiple linear regression (UAV-MLR) model was developed using ΔDSMt, NDVI, GNDVI, and NGRDI to integrate complementary structural and spectral information. Evaluated on the calibration dataset, the UAV-MLR model demonstrated high fitting performance (adjusted R2 = 0.9858, RMSE = 14.95 cm, MAE = 11.31 cm), outperforming the ground-based simple linear regression (G-SLR) model based on stem diameter (adjusted R2 = 0.9615, RMSE = 39.68 cm, MAE = 29.16 cm). By integrating structural and spectral information, the proposed approach reduced RMSE by 62.3%, offering a highly accurate, non-destructive tool for crop monitoring and precision agriculture.
Benoit DE SELON · Jean‐Eudes Hollebecq · Jordan Bernigaud-Samatan · Caroline Chaynes · Solenne Faul-Godec · Mathieu MARGUERIE · Afonso Ponce · Mario Serouart · Samuel Thomas · Tania ROUGIER
CountingSegmentation
The characterization of plants and their environment is changing scale with the development of digital plant phenotyping tools and methods. Phenotyping and envirotyping tools are more accessible, enabling multi-scale observation of plants and their environment—even molecules—through multi-plot observation via satellite. At the same time, large amounts of data are more accessible and easier to exploit through indexing in open information systems. Processing is facilitated by an increasing number of sophisticated tools, particularly with the help of artificial intelligence, thus facilitating analysis by the operator. Two applicated examples are detailled here regarding segmentation for combined mixed crops and insects identification and counting. Environmental and phenotypic data can be combined in predictive or decision support models. These data are valuable assets for accelerating the deployment of agroecology and resilient, sustainable agriculture.
Abstract Rain-induced cracking is among the most damaging preharvest disorders of sweet cherry because even narrow cuticular fractures reduce fresh-market value and create entry sites for water loss and fungal infection. Conventional inspection is subjective and is least reliable for low-contrast microcracks near the pedicel cavity. This study developed a Fourier-guided multi-scale Vision Transformer (FGM-ViT) for joint fruit-level severity grading and pixel-level crack segmentation. The dataset comprised 960 individual fruit from four commercial cultivars and eight orchard blocks, imaged from four rotational views after natural rainfall exposure or controlled water-immersion challenge. Crack severity was assigned as intact, microcrack, moderate, or severe using stereomicroscopic measurements of cumulative crack length, maximum width, affected surface area, and crack location. All views of an individual fruit and all observations from the same orchard block were constrained to the same fold in a nested five-fold grouped evaluation. FGM-ViT combined a multi-scale spatial encoder with luminance-normalised Fourier residual tokens and frequency-guided cross-scale attention. The complete model achieved 93.8 ± 0.9% fruit-level accuracy, 93.7 ± 1.0% macro-F1, and 0.842 ± 0.018 Dice coefficient for crack segmentation. It exceeded the spatial-only transformer by 2.2 percentage points in accuracy and 2.2 points in macro-F1, with the largest gain observed for microcracks. Classification errors were restricted almost entirely to adjacent severity categories. Under pedicel overlap, glare, brightness shifts, and motion blur, FGM-ViT retained a 2.7–3.5-point advantage over the spatial-only model. Frequency-band occlusion indicated that mid-to-high spatial frequencies carried complementary evidence for narrow fractures, whereas spatial features remained essential for crack location and marketability interpretation. The dual-output framework provides an auditable, low-cost RGB approach for sweet-cherry sorting, cultivar screening, and rain-cracking phenotyping.
Background and Aims: Pollen germination and tube growth are critical stages of plant reproduction that are highly sensitive to temperature but remain labor-intensive to quantify. This study aimed to develop a deep learning-based approach for pollen phenotyping and to characterize the temperature dependence of pollen germination and elongation in Douglas-fir (Pseudotsuga menziesii). Methods: A convolutional neural network (CNN) was trained to segment pollen grains and classify germination status from microscopic images. Germination percentage and pollen length were quantified across a range of incubation temperatures from 5 to 40°C. Gamma functions were fitted to temperature response curves to estimate the optimal temperatures for pollen germination and elongation among Douglas-fir populations collected across an elevational gradient. Key Results: The CNN achieved high segmentation accuracy (intersection over union = 0.846), accurately distinguishing pollen grains from the background but showing moderate accuracy in separating germinated from ungerminated pollen during early elongation. Both pollen germination and elongation exhibited bell-shaped temperature response curves with distinct thermal optima. Pollen elongation consistently reached its optimum at higher temperatures than pollen germination. Estimated optimal temperatures fell within a narrow range, approximately 19 to 23°C. No significant relationship was detected between elevation and thermal optima, although some higher-elevation populations exhibited lower optimal temperatures. Comparisons with previous analyses of three western North American conifers showed that each species occupied a distinct reproductive thermal niche corresponding to the spring temperatures of its native habitat. Conclusions: Deep learning provides an efficient approach for high-throughput quantification of pollen germination and elongation from microscopic images. The narrow thermal range for reproductive performance suggests that Douglas-fir pollen is sensitive to temperature variation and that warming climates may alter reproductive success. These findings improve our understanding of the thermal sensitivity of conifer reproduction and provide a scalable framework for assessing impacts of climate warming on forest regeneration.
Plant cells grow and differentiate in an ever-changing environment characterized by transient signals and stimuli. The ability of plant cells to perceive, integrate, and dynamically respond to these stimuli underpins a plant adaptation and survival. Among the plethora of complex stimuli plant cells are exposed to, several stimuli have a mechanical component, for example, wind, touch, contact with insects, penetration of pathogens, and even intrinsic mechanical stresses arising during tissue growth. Despite the presence of a load-bearing cell wall that separates cells from the environment and fixes their location within a tissue, plant cells are responsive to mechanical stimuli. However, several questions remain unanswered around how mechanical stimuli are perceived and translated into cellular responses. Here we establish a system enabling application of quantifiable localized mechanical stress while simultaneously capturing cellular responses with high spatio-temporal resolution using confocal imaging. We show that this system enables estimation of locally applied pressure and provides access to the temporal and spatial details of subcellular events in intact living tissues upon touch. We propose that observing these subcellular events at high spatiotemporal resolution and linking their dynamics to the intensity of mechanical stimuli may uncover the molecular mechanisms underlying plant cell responses to touch.
Reproduction assets foundThe paper's quantitative analysis code is publicly available: the authors state that R scripts for load-cell force readout analysis and Fiji/R scripts for fluorescence signal profiling are available at the Microindentation-toolbox GitHub repository, and the LabView load-cell readout software (Flamos) used for the paperCode · publicpush on
134
the mobile plate of the load cell. The needle was then lifted and lowered again after a
135
few seconds interval to create repeated touches on the load cell mobile plate. Force
136
readouts from the software were then analyzed using an R (R Core Team, 2025) script
137
(publicly available at the Github repository
138
https://github.com/AnnalisaBe/Microindentation-toolbox.git). For pressure estimates,
139
the plant-needle interface was approximated to the curved surface of a half sphere.
140
The pressure applied on the tissue upon needle movement was then estimated as the
141
force exerted by the needle movement (as calibrated by the load cell) divided by the
142
curved surface of Open asset ↗AnnalisaBe/Microindentation-toolboxpdf-raw-page:4 lines:1-88Plant phenotyping relevance matchbioRxiv · Europe PMC · checked 15 Sept 2026
Leaves of terrestrial plants possess heterogeneous anatomical structures composed of multiple cell layers. Although biochemical differences among leaf tissues have been inferred from protein analyses and anatomical studies, direct comparisons of dynamic photochemical responses among tissues while preserving their spatial context remain challenging. Leaves experience intrinsically heterogeneous environments because incident light enters primarily from above and propagates through complex internal leaf structures. Therefore, analyzing photosynthetic activity at the tissue and cellular levels is essential for understanding how photosynthesis operates within structurally heterogeneous leaves. Here, we combined live leaf-section imaging with microscopic Imaging-PAM chlorophyll fluorescence measurements to analyze photochemical responses at the tissue level in living leaf sections. In dorsiventral dicot leaves, palisade tissues exhibited a higher effective PSII quantum yield [Y(II)] and more rapid induction of regulated energy dissipation [Y(NPQ)] than spongy tissues, indicating higher photosynthetic capacity and photoprotective activity. In contrast, rice leaves, which lack palisade-spongy differentiation, showed uniform photochemical responses along the adaxial-abaxial axis. In the C4 plant finger millet, mesophyll and bundle sheath cells displayed distinct photochemical responses consistent with their functional differentiation in C4 photosynthesis. These results demonstrate that photosynthetic responses are spatially organized according to leaf anatomical structure. Although section-based measurements do not reproduce the native optical and gas environments of intact leaves, they enable the comparison of intrinsic tissue-specific photochemical properties under approximately equivalent illumination. The present approach enables tissue-resolved chlorophyll fluorescence analysis in living leaves, providing a new framework for investigating how leaf architecture shapes the spatial organization of photosynthetic activity.
This study presents a systematic quantitative multi-elemental investigation of four major plant organs (roots, stems, leaves, and flowers) of Nerium oleander using calibration-based Laser-Induced Breakdown Spectroscopy (LIBS), with the results validated by Atomic Absorption Spectroscopy (AAS). Plasma characterization was carried out using Boltzmann plot and Stark broadening analyses, while negligible self-absorption observed through the Hα emission line confirmed optically thin plasma conditions and reliable quantitative measurements. A total of nine elements were detected, including Fe, Zn, Mn, Ca, Mg, K, Na, Cu, and Ni, each exhibiting different concentration levels across the analyzed tissues. Compositional analysis using standard calibration curve-based LIBS demonstrated that elemental concentrations were non-uniform, showing marked variations between the different plant tissues. Among the detected elements, calcium emerged as the most prevalent across all tissues. The highest calcium concentration was observed in leaves (16,385 mg L-1), followed by roots (12,092 mg L-1) and flowers (11,185 mg L-1). Root tissues exhibited elevated concentrations of Fe and Mn, reaching 1918 and 513 mg L-1, respectively. In contrast, flowers showed the highest Mn concentration (1888 mg L-1), while leaves were enriched in Mg (4445 mg L-1) and K (3700 mg L-1). The highest Na concentration was observed in stems (8025 mg L-1). Trace metals, including Cu, Zn, and Ni, were detected at comparatively low concentrations, while Pb remained undetected in all samples. The strong agreement between LIBS and AAS measurements confirms the reliability of the proposed methodology and demonstrates the potential of LIBS as a rapid and non-destructive tool for elemental assessment of medicinal plants.
Many labor-intensive tasks in fruit production such as pruning require physically interacting with the plant (e.g., pushing, pulling, bending limbs, etc.). Due to increasing labor shortages, there is widespread interest in the adoption of robotics in this area. When the robot must physically interact with the system, a deformable model of the plant is beneficial. Coupled, spring-loaded beams have been used in prior work to build deformable plant models in simulation for learning, planning, and control, but this approach has rarely been validated against the mechanical properties of real-world plants. In this paper, we use this rigid-body model, grounded in real material properties and beam bending theory, to simulate the deformation of blueberry canes under loading. To validate the model, we simulate the canes in MuJoCo and compare their behavior with real-world data collected from probing canes at a commercial farm with a custom testbed. We perform sensitivity analysis on several key modeling variables and show that this approach is highly sensitive to the plant diameter calculations and a priori flexural modulus estimation, which is dependent on season and blueberry variety. We also share our dataset of live blueberry cane deformation, including RGB-D images and measured forces and displacements.
Reproduction assets foundThe paper shares its in-field blueberry cane deformation dataset (RGB-D images, measured forces and displacements) via a figshare link provided in a footnote. Other allowed URLs (CloudCompare, SALib references) are generic tools/citations, not paper-specific assets.Dataset · publicionally, we performed sensitivity analysis on several modeling variables to identify potential significant sources of error.
In summary, the contributions of this paper are:
•
A dataset capturing live blueberry cane deformation (visual deformation under load, plus measured forces and displacements) 1 1
1
Data can be accessed at https://figshare.com/s/f82e41c7115187e42df8 .
•
Evaluation of the physics model against real-world data collected from a commercial blueberry farm with a custom testbed.
•
Sensitivity analysis of several variables for lumped-parameter rigid-body models.
This paper explores the impact of season, age, and variety on the flexural modulus of blueberry plants, and its impaOpen asset ↗figsharelines:63-82Plant phenotyping relevance matchCrossref · checked 15 Sept 2026
Classification2D/3D reconstructionWater status / transpiration
Abstract Monitoring pollen is important for crop productivity, plant breeding, and environmental forecasting, yet current approaches remain limited in throughput and physiological insight. Here, we combine single-shot plasma emission imaging with artificial intelligence for pollen morphology reconstruction and apparent hydration state classification. Plasma plumes generated during laser–pollen interaction provide indirect measurements of pollen morphology, extending plasma-based analysis beyond conventional spectroscopic readouts. A conditional generative adversarial neural network reconstructed morphology from such plasma images, while chromaticity analysis revealed systematic color shifts between hydrated and dehydrated grains. A support vector machine using chromaticity histogram and scalar color features distinguished these groups with ∼76% leave-one-out accuracy. The results suggest that plasma emission contains structural and hydration-associated information, supporting rapid pollen morphology inference and physiological phenotyping for agricultural and environmental monitoring.
Abstract Developmental progression in plants is inherently a time-to-event process, yet plant phenology data are often analyzed as ordinary endpoint traits even when some individuals fail to reach the target stage within the observation window. Such observations are right-censored rather than missing and should be retained in inference. Here, we evaluated 21 soybean cultivars across 10 controlled-environment treatment combinations varying in CO 2 concentration, photoperiod, and LED light quality, and analyzed days from sowing to the R6 stage using a practical workflow that combined Kaplan-Meier survival analysis with a Bayesian Weibull accelerated failure time (AFT) model. Based on 210 cultivar × treatment observations derived from 1,050 plants, treatments resolved into favorable, intermediate, and strongly inhibitory classes. Elevated CO 2 showed the strongest association with reproductive timing: increasing CO 2 from ambient to 1000–1400 ppm reduced median survival time, while the additional gain from 1000 ppm to 1400 ppm was minimal, indicating a saturating response. Blue - Red spectral treatments and 6–8 h photoperiods were associated with rapid and synchronized development within the tested treatment combinations. The Bayesian AFT model included treatment as a random effect to account for the chamber-level experimental structure; the between-treatment standard deviation was estimated at 0.02 (95% credible interval: 0.00–0.06), indicating negligible chamber-to-chamber variation. The highest-ranked predicted combination (1400 ppm, blue: red = 2:1, 6 h) gave a median time to R6 of 66.90 d (mean 72.18 d). This study provides a practical and transferable workflow for censored plant phenology datasets in controlled-environment research, phenotyping, and breeding. The workflow integrates established survival analysis methods and explicitly accounts for chamber-level design, offering a framework for structured plant phenotyping experiments.
Soil contamination represents a growing environmental concern, particularly in areas such as the "Land of Fires" (Campania region, Italy). This study evaluates the potential of Zea mays L. as a bioindicator of soil contamination through a multiscale approach. In bin-like containers under natural conditions, maize plants were grown on untreated (C) and artificially contaminated (T) soil with a mixture of heavy metals (Pb, Zn, Cr) and benzo[a]pyrene. Analyses were conducted at leaf, canopy, and field scales, integrating eco-physiological, spectral, and UAV measurements. At the leaf level, contamination induced functional alterations in photochemical and spectral traits that were strongly dependent on the growth stage, being most pronounced early in the season and progressively attenuating thereafter. At the canopy level, contaminated plants developed less height and leaf area, together with lower chlorophyll- and canopy-density-sensitive spectral indices, consistent with reduced vegetative vigor under contamination stress. Solar Induced Fluorescence (SIF) showed no residual difference between treatments once normalized for biomass production at the end of the growing season, in agreement with leaf-level active fluorescence measurements taken on the same date. At the field scale, UAV-based classification achieved superior performance with hyperspectral relative to multispectral data in distinguishing contaminated areas. Taken together, these results show that, under the experimental exposure conditions adopted here, the detectability of soil pollution by maize progressively shifts from early, transient physiological signals to persistent structural differences, while UAV hyperspectral data provided the highest discrimination between treatments. By tracking how leaf-level functional alterations translate into canopy- and field-scale structural separability, this study provides a scale-explicit assessment of maize as a promising bioindicator of soil contamination.
Accurate point-cloud segmentation of cotton organs is essential for precise phenotypic characterization. However, reliable instance segmentation of cotton bolls in field-derived point clouds remains challenging because foliage occlusion, contact between adjacent bolls and incomplete reconstruction obscure instance boundaries. Here we present RQ-PointNeXt, an end-to-end framework that directly maps input point clouds to boll instance masks within a unified trainable network. Built on PointNeXt, it incorporates relative-elevation geometric channel attention in the shallow encoder to fuse global channel context with elevation and surface-normal cues. Its Query–mask branch integrates semantic guidance, center-seeded queries and a center-aware mask prior to suppress background responses, localize instances and constrain mask extent. Hungarian matching and multitask optimization establish one-to-one query–instance assignments, whereas query-based decoding produces instance masks without external geometric clustering. We evaluated the framework on 226 field-grown cotton plants containing 720 annotated boll instances reconstructed from UAV multi-view imagery using neural radiance fields. On the held-out test set, overall accuracy, mean class accuracy and mean intersection over union reached 0.8942, 0.8980 and 0.8076, respectively. AP25, AP50 and AP75 were 0.7359, 0.5585 and 0.2777, yielding an mAP25/50/75 of 0.5240. The framework provides instance-level outputs for boll counting and spatial analysis in high-throughput field phenotyping.
Reproduction assets foundThe paper's own field cotton point-cloud dataset (226 plants, 720 annotated bolls) is only available upon request. However, the authors directly used the public UGA-BSAIL Cotton Plants with Foliage point-cloud dataset (with their added boll instance annotations) as an evaluation benchmark for RQ-PointNeXt, and it is公开发Dataset · publict to the pointwise overlap between
predicted and ground-truth instances.
To further evaluate the proposed method under conditions of relatively high point-
cloud completeness, experiments were conducted using the public UGA-BSAIL Cot-
ton Plants with Foliage dataset. The point-cloud data are publicly available through
Figshare (https://figshare.com/projects/Cotton_plant_with_foliage/258065, accessed on
8 September 2026), while the associated code and documentation are hosted on GitHub
(https://github.com/UGA-BSAIL/Cotton_plants_with_foliage, accessed on 8 September
2026). The dataset contains relatively complete cotton plant point clouds, surface-normal
attributes, and semantic labels distinOpen asset ↗Figshare · Cotton_plant_with_foliage/258065pdf-raw-page:16 lines:1-52Code / dataset availability confirmedEurope PMC · checked 15 Sept 2026
Kawai T, Teramoto S, Ma X, Fukushima D, Hmwe KK, Kimani SM, Tokida T, Uga Y.
RiceMultimodalX-ray / CTRootMorphology / geometry measurementPhysiological trait estimationGrowth / time-series analysisGrowth / development / phenologyRoot system architecture
Rhizosphere oxidation is a key adaptive mechanism in reductive soil environments, in which oxygen released from roots alters rhizosphere redox conditions and regulates biogeochemical processes. Rice plants possess an internal oxygen transport system, and radial oxygen loss (ROL) from roots is closely associated with root development. However, the spatial patterns of ROL in soil and their relationships with root traits remain poorly characterized. In this study, we developed a multimodal imaging system that integrates planar oxygen optodes with X-ray computed tomography to simultaneously visualize rhizosphere oxidation and root development in rice. Daily time-course tracking of individual crown roots revealed dynamic changes in the spatial distribution and magnitude of rhizosphere oxygen in relation to root elongation and aging. Root thickness was positively correlated with dissolved oxygen levels near root tips. Genotypic comparisons further identified a cultivar with reduced rhizosphere oxidation despite possessing thicker roots among the tested genotypes, thereby indicating the involvement of additional physiological processes. Overall, these findings demonstrate that rhizosphere oxidation is regulated by root growth stage and thickness and dynamically modulated during root development.
Reproduction assets foundThe paper's Data availability statement explicitly deposits the authors' RG2DO-Root analysis program together with sample optode and CT images (the paper's phenotyping inputs) in a public GitHub repository, matching the allowed URL.Code · publicing
8
This work was supported by project JPNP18016, commissioned by the New Energy and
9
Industrial Technology Development Organization (NEDO), JST CREST (JPMJCR17O1),
10
and JST ALCA-Next (JPMJAN23D3).
11
12
Data availability
13
The source code and sample data (optode and CT images) are available from the
14
GitHub repository (https://github.com/tsubasa-kawai28/RG2DO-Root).15
16
References
17
Aguilar EA et al. 2003. Oxygen distribution and movement, respiration and nutrient
18
loading in banana roots (Musa spp. L.) subjected to aerated and oxygen-depleted
19
environments. Plant Soil. 253:91–102. https://doi.org/10.1023/A:1024598319404.20
Armstrong W, Wright EJ. 1975. Radial oxygen loss fromOpen asset ↗https://github.com/tsubasa-kawai28/RG2DO-Root · RG2DO-Rootpdf-raw-page:19 lines:1-82Plant phenotyping relevance matchCrossref · checked 15 Sept 2026
Geospatial artificial intelligence (GeoAI) integrates multimodal remote sensing with deep learning to model complex agricultural systems at scale. Within this framework, accurate and non-destructive prediction of seed composition from in-season standing crops is essential for breeding and precision agriculture. This study developed an end-to-end convolutional neural network (CNN) framework to estimate eight seed traits (protein, oil, sucrose, fiber, starch, ash, complex and simple carbohydrates) from UAV-based multisensor imagery and associated genotype and phenological metadata. A total of 372 soybean samples were collected over two growing seasons (2020–2021) from two fields in Missouri, with UAV flights capturing multispectral (MSI), thermal (THR), and LiDAR (LDR) data at four time points spanning vegetative to reproductive growth stages. CNN models were trained in single- and multi-date configurations, incorporating genotype (GEN) and days after sowing (DAS) as additional features. The highest accuracy was achieved for sucrose (R2 = 0.80), followed by simple carbohydrate (R2 = 0.70) and starch (R2 = 0.55), with notable gains from GEN and DAS. Multi-date models incorporating earlier acquisitions often matched or outperformed later or all-date combinations. Among modalities, MSI provided the most robust estimates, with limited added value from LDR or THR. Unlike feature-based pipelines prone to multicollinearity, this image-to-trait approach enables automated, scalable prediction of soybean seed composition for in-season, field-level assessment.
Lois Onyejere Nwobodo · Chinonso J. Okonkwo · Edith Angela Ugwu · Udeh Chukwuma Callistus · Okorie Kingsley Maduabuchi · Ngene John Ndubisi · Agbo Kenechukwu Martin
Field / plotLeafClassificationStress / disease detectionDisease symptoms / severity
Plant diseases threaten agricultural productivity, and automated image analysis can support early identification of visible disease symptoms. This study introduces HybOptic-CNN, a convolutional neural network (CNN) whose learning rate and batch size are selected using a hybrid Whale Optimization Algorithm--Grey Wolf Optimizer (WOA-GWO). Nine disease classes were selected from the 22-class CCMT field-image dataset, and 152 local farm leaf images were collected in Enugu State, Nigeria. Of the local images, 122 (80.3%) were added to the model-development data for training and validation, whereas 30 (19.7%) formed an independent Nigerian hold-out set excluded from augmentation, class balancing, early stopping, validation, and hyperparameter selection. Across 10 model-development runs, the optimized model achieved 96.8 ± 0.4% mean validation accuracy, 95.2 ± 0.5% macro-precision, 94.9 ± 0.6% macro-recall, and 95.0 ± 0.5% macro-F1, compared with 90.3 ± 0.9% validation accuracy and 86.3 ± 1.2% macro-F1 for the baseline. The optimized model improved mean validation accuracy by 6.5 percentage points and converged 14.6 epochs earlier. On the independent 30-image Nigerian hold-out, HybOptic-CNN achieved 93.3% accuracy and 93.1% macro-F1 across four represented disease classes. A web application integrating the trained classifier was also demonstrated. These results support improved model-development performance through hybrid hyperparameter selection and motivate broader multi-location field evaluation.
Reproduction assets foundThe paper's Data availability statement points to two public sources: a Mendeley dataset (the locally collected Nigerian field images) and the Kaggle CCMT plant disease dataset used as the principal image source. Only the Kaggle URL matches an allowed URL; the Mendeley URL is not in the allowed list, so only the CCMT/KDataset · publicnt and independent field-test data and should pub-
lish the class-wise split manifest, random seeds, WOA-GWO
numerical settings, and evaluation code so that the reported pro-
cedure can be reproduced and extended.
Data availability
The data used in this study are available at https://
data.mendeley.com/datasets/bwh3zbpkpv/1 and https://www.kaggle.com/datasets/rahimanshu/ccmt-plant-disease-dataset.Declaration of competing interest
The authors declare that they have no known competing fi-
nancial interests or personal relationships that could have ap-
peared to influence the work reported in this manuscript.
Funding
The authors received no specific funding from any public,
commercial, or not-fOpen asset ↗Kaggle · rahimanshu/ccmt-plant-disease-datasetpdf-raw-page:12 lines:1-78Plant phenotyping relevance matchCrossref · checked 15 Sept 2026
The precise temporal characterization of early growth dynamics in crops under abiotic stress is critical for stress resistance management in smart agriculture. However, traditional manual measurement methods struggle to achieve high-throughput and accurate quantification of multiple phenotypic parameters over continuous time series. To address this, we developed a full-time-series crop growth monitoring system that continuously collects fine-grained image data of cucumber seed germination and seedling growth under soil culture conditions, constructing an annotated dataset for identifying germination status and quantifying cotyledon area. The YOLOv8n-F_SGX and YOLOv8-seg models were developed and deployed, achieving detection and segmentation accuracies of 0.978 and 0.987, respectively, which enabled automatic monitoring of germination rate and precise extraction of cotyledon area. To further investigate the mitigating effects of nanomaterial priming on salt stress, we conducted germination and seedling growth experiments on cucumber seeds using a series of ZnO-NPs (particle size 30 nm) suspension concentrations at different salt stress levels. The results demonstrated that under salt stress conditions of 0–150 mmol·L −1 , priming with 100 mg·L −1 ZnO-NPs resulted in optimal germination dynamics (highest germination rate and fastest germination speed), along with the largest cotyledon area and highest growth rate in seedlings. In contrast, when the ZnO-NPs concentration increased to 400 mg·L −1 or higher, it significantly inhibited seed germination and seedling growth.
Volatile organic compounds (VOCs) released by plant leaves play key roles in stress signaling, plant-atmosphere interactions, and plant defense. Among these, wound-induced VOCs (wVOCs) are emitted within seconds of mechanical damage, herbivory, or environmental disturbance. Their emission dynamics depend strongly on the timing, severity, and method of tissue disruption. Yet, accurate quantification remains challenging due to mechanical artifacts, variable exposure conditions, and delays between injury and measurement. This study presents a standardized within-chamber leaf excision protocol for real-time monitoring of wVOCs and gas exchange. A surgical-grade cutter was integrated into a portable gas-exchange chamber to enable clean, controlled cuts within a sealed chamber under stable light, humidity, CO2, and temperature conditions. A proton-transfer-reaction time-of-flight mass spectrometer (PTR-TOF-MS) continuously measured volatile emissions at the chamber outlet, minimizing delay and signal distortion. This setup resolves emission onset, peak timing, maximum rise rate, and total release with high temporal fidelity. Application of the method to Quercus rubra, Acer platanoides, and Gossypium hirsutum demonstrated its ability to resolve distinct wound-induced emission patterns across contrasting leaf types. By eliminating delays associated with conventional sampling, this method resolves the full kinetic trajectory of wound-induced emissions and overcomes major limitations of previous approaches. It provides a robust framework for studying rapid stress responses in plant physiology, ecological biochemistry, and plant-atmosphere interactions.
Transmission electron microscopy (TEM) is widely used to examine plant cellular ultrastructure, but sample preparation remains challenging because polysaccharide-rich cell walls, large vacuoles and complex membrane systems compromise staining and structural preservation. Here, we developed a plant-adapted TEM preparation method based on a modified osmium-thiocarbohydrazide-osmium (OTO) staining workflow combined with optimized washing, dehydration and resin infiltration. Using tomato roots and leaves, we show that the optimized method produces cleaner backgrounds, better-defined cellular boundaries and improved preservation of cellular morphology compared with conventional preparation. Fine membrane-associated structures, including mitochondrial cristae, chloroplast grana and stroma lamellae, nuclear membranes and other endomembrane structures, were more clearly resolved, while staining-related precipitates and diffuse background artifacts were reduced. This method provides a practical approach for high-contrast TEM imaging of plant tissues and ultrastructural analysis of plant cells and organelles.
Accurate and dynamic monitoring of wheat phenotypes is essential for breeding decision-making and crop management. However, RGB image-based phenotyping still suffers from expensive pixel-level annotation, unstable organ-level segmentation across growth stages, and limited multi-trait extraction under complex field conditions. To address these issues, a high-throughput phenotyping framework (WheatScoper) was proposed, enabling organ-level segmentation and plot-level multi-trait extraction. To reduce annotation cost, a structure-aware geometry-assisted annotation (SAGA) algorithm was developed, yielding an approximately 7.6-fold improvement in annotation efficiency over fully manual annotation. To enable efficient organ-level segmentation, a lightweight semantic segmentation network (WheatScopeNet) was developed by integrating parallel hybrid spatial modeling with cross-scale feature fusion. On the held-out test set from the same site and growing season, WheatScopeNet achieved an mIoU of 0.869 and an mDice of 0.930. Leveraging the segmentation results, an automated system was established to extract 41 multi-view image-derived traits (I-traits) across four core phenotypic dimensions. The extracted I-traits supported the estimation of eight manually measured agronomic traits, with R 2 values ranging from 0.477 to 0.697. Notably, the correlations between stay-green-related-traits and yield varied distinctly with viewing-position. Only upper side-view indicators remained significantly correlated with yield, with Side-up final GPAR showing the strongest association, whereas top-view GPAR-derived indicators showed weak associations. Finally, a web-based platform integrating cascaded inference, segmentation visualization, and automatic I-trait extraction was developed. The platform provides an end-to-end solution for field wheat phenotyping and supports breeding decision-making and crop management.
Reliable plant segmentation in high-throughput phenotyping must transfer across species and imaging conditions without repeated model tuning or extensive reannotation. We compare three segmentation strategies using images from Oak Ridge National Laboratory's Advanced Plant Phenotyping Laboratory: (i) fixed color-based thresholding, (ii) supervised U-Nets trained from scratch, and (iii) pretrained vision transformers fine-tuned for binary segmentation. Models were evaluated on a held-out test set and a generalization set that comprised unseen species. On the held-out test set, thresholding, the best U-Net, and the best vision transformer achieved mean Dice scores of 58.3, 96.6, and 97.3, respectively. On the generalization set, the corresponding Dice scores were 56.5, 86.2, and 95.7. Thresholding remained effective on some datasets but failed when plant appearance changed. Supervised U-Net training resolved within-distribution errors but failed to generalize to novel species and backgrounds. Pretrained vision transformers consistently produced high-accuracy segmentations across the evaluated species, views, soil backgrounds, and tray types. These results benchmark the practical progression from fixed rules to task-specific supervision and pretrained visual representations for controlled-environment plant phenotyping.
High-throughput acquisition of crop phenotypic information is one of the key technologies for achieving intelligent facility agriculture and precision breeding. Traditional phenotypic data collection methods suffer from low efficiency and strong subjectivity, making it difficult to achieve multi-scale continuous monitoring and meet the demands of modern research and production. This paper systematically reviews the technological framework and development trajectory of optical sensing technology-driven phenotypic platforms for facility crops. First, starting from optical sensing technologies, a comparative analysis highlights the advantages and limitations of RGB, multi-/hyperspectral, thermal infrared, and LiDAR sensors in phenotypic perception. Second, the characteristics and applicable scenarios of stationary, rail-mounted, mobile robot, and unmanned aerial vehicle (UAV) platform architectures are summarized. Furthermore, the evolution of phenotypic data processing methods is examined, focusing on the shift from traditional feature engineering to deep learning-driven approaches. Finally, key challenges such as multimodal data fusion, system cost, and real-time performance are discussed, along with the future direction of phenotypic platforms toward intelligent closed-loop decision-making systems. This article systematically reviews the facility agriculture phenotyping platforms driven by optical sensing technology, and also incorporates representative research progress in field phenotyping studies. These advances provide transferable sensing technologies, methodological frameworks, and platform design concepts that can facilitate the development of phenotyping platforms for controlled-environment agriculture.
Abstract Background : Seed size, shape, and colour are key determinants of Bambara groundnut (BGN) seed quality, processing efficiency, and market value. However, conventional assessment methods are slow, subjective, and poorly suited to large-scale characterisation. There is a need for standardised, non-destructive, high-throughput phenotyping methods to support breeding, seed quality assessment, and germplasm evaluation for this resilient yet underutilised crop. Methods : Videometer multispectral imaging (MSI) was used to generate reproducible seed morphometric descriptors for 106 BGN accessions. Twenty seed traits were quantified, including size (area, length, width, perimeter, and volume), shape indices (rectangularity, eccentricity, compactness, form factor, pointness, beta shape, diameter area, width of blob end, and compactness ellipse), and colour metrics (hue, saturation, lightness, and CIELab-A1). Data were analysed using analysis of variance, agglomerative hierarchical clustering, and principal component analysis to assess phenotypic variation. Results : Significant genotypic effects were detected across all measured traits (p < 0.05), indicating substantial phenotypic diversity among the 106 accessions. Seed size and shape traits exhibited wide variation (area ≈ 35–148 mm 2 ; length ≈ 7.3–15.5 mm; and width ≈ 6.5–12.7 mm). Multivariate analyses identified exploratory, dataset-specific phenotypic groupings that broadly reflected relative differences in seed size, shape, and colour among accessions, supporting germplasm characterisation and the identification of contrasting parental material for pre-breeding. Conclusions : MSI-based phenotyping provides a rapid, non-destructive, and high-throughput approach for quantifying seed morphometric variation in BGN. The generated seed trait data provide a standardised basis for germplasm characterisation, the identification of contrasting parental material, and early-stage screening, establishing MSI as a valuable phenotyping foundation for future BGN improvement.
Reliable prediction and interpretation of photosynthetic rate are important for precision environmental management in greenhouse tomato production, but model stability is often limited by variable redundancy, measurement-session effects, and environmental heterogeneity. This study developed an integrated gas-exchange-data-based framework for key-factor selection, photosynthetic-rate reconstruction, cross-session validation, environmental correction, and physiology-informed limitation screening. Core predictors were first identified from high-dimensional gas-exchange variables using K-means clustering and random-forest importance analysis. A CatBoost–ExtraTrees–RBF-SVR stacked ensemble was then constructed to reconstruct the leaf photosynthetic rate from selected gas-exchange variables, and its cross-session generalization was evaluated using nested cross-validation and leave-one-file-out (LOFO) extrapolation. Environmental correction was further applied to improve cross-session comparability, and rule-based limitation screening was used to classify potential environmental constraints associated with reduced photosynthesis. The stacked model achieved an RMSE of 0.806 and an R2 of 0.9918 under nested cross-validation, and an RMSE of 0.814 and an R2 of 0.9917 under LOFO extrapolation. After environmental correction, the cross-session variability of the environmentally corrected photosynthetic indicator was reduced by 96.0%, reflecting improved cross-session comparability. Under deployable sensor inputs, performance decreased markedly (A1: RMSE = 4.578, R2 = 0.735; A2: RMSE = 4.610, R2 = 0.731), compared with the gas-exchange baseline (RMSE = 0.833, R2 = 0.991). Thus, routine greenhouse sensor models should be regarded as approximate rather than high-accuracy substitutes for the gas-exchange-based model.
Reproduction assets foundThe paper analyzes a publicly available greenhouse tomato gas-exchange dataset (Manjarrez-Sanchez & Martinez-Carrillo, Data Brief 2020), which is the exact phenotype/sensor input data for this study's photosynthesis reconstruction analysis. The authors' own processed data and analysis scripts are only available upon请求,Dataset · publicThe data used in this study were obtained from the publicly available greenhouse
tomato gas-exchange dataset reported by Manjarrez-Sanchez and Martinez-Carrillo [29].Open asset ↗pdf-raw-page:4 lines:1-38Plant phenotyping relevance matchOpenAlex · checked 15 Sept 2026
Quantitative measurements of insect flight behaviour are essential for understanding the biomechanics, control, and ecology of flight, yet obtaining such measurements under free-flight conditions remains challenging. Small body size, rapid wing motion, visual symmetry, and frequent occlusions complicate three-dimensional pose estimation, often requiring restrictive experimental setups or substantial manual annotation. We present FLiTrak3D, an open-source Python package for estimating insect flight kinematics from multi-view videography. It combines machine-learning-based markerless tracking with biomechanical modelling to reconstruct and parametrise insect body and wing motion. The workflow integrates image preprocessing, including dynamic image cropping and enhancement, two-dimensional bodypart localisation, three-dimensional reconstruction, and optimisation-based skeletal fitting. A key innovation is the use of spatio-temporal encoding across synchronised camera views and adjacent frames to improve neural-network awareness of spatial and temporal context during bodypart localisation. Multi-view image stitching allows the network to use cross-view spatial relationships, reducing left-right bodypart misidentifications, while temporal encoding stacks consecutive greyscale frames into RGB images provide short-term motion information. A species-specific skeleton is then fitted to reconstructed keypoints to enforce kinematic constraints, and estimate body and wing orientations. We demonstrate the approach using free-flying Aedes aegypti mosquitoes, achieving bodypart localisation errors close to human labelling, and realistic flight kinematics.
Zea mays (maize) is a critical global crop, generating more than 1 billion metric tons of grain annually. Plant morphology is tightly associated with final yield and is determined by the process of development. Accurate assessment and recording of mature plant morphology is essential for linking genotype to phenotype and for assessing environmental influences on plant development and yield. In maize, imaging of fully emerged aerial organs provides an accessible and informative developmental end point for comparative phenotypic analyses. Here, we provide a protocol for capturing high-quality images of mature maize plants and detached leaf blades, suitable for publication and downstream quantitative trait extraction. We emphasize standardized imaging conditions, including high-contrast backgrounds, diffuse lighting, fixed camera positioning, and consistent scale calibration, to minimize technical variation and to facilitate automated image processing. Guidance is provided for both laboratory- and field-based imaging setups and uses widely available equipment. Images generated using this protocol can be analyzed using platforms such as FIJI to extract morphometric traits, including length, width, and outline-based shape descriptors. By prioritizing standardization at the point of acquisition, this protocol supports reproducible, scalable phenotyping of mature maize morphology for developmental studies in maize.
In plants, final organ shape and position influence function and whole plant productivity. In Zea mays (maize), leaf, ear, and tassel morphology are all key agronomic features influencing yield. These aerial structures originate from stem cell populations within meristems. Understanding how meristems are regulated and how stem cells differentiate over developmental time to form mature leaves, ears, and tassels is important for identifying novel routes for maize improvement. High-resolution imaging of meristem morphology is a key component of developmental studies. Scanning electron microscopy (SEM) is widely used to visualize meristem topology; however, conventional SEM preparation requires fixation, dehydration, drying, and conductive coating. These steps are time-consuming, technically demanding, and can introduce damage. Here, we present a rapid protocol for imaging freshly dissected meristems from maize using a benchtop scanning electron microscope. Fresh tissues are mounted directly onto conductive adhesive and imaged under high vacuum at low accelerating voltage, enabling high-quality micrographs to be captured within minutes of dissection. Although dehydration and tissue collapse can occur under vacuum, rapid sample handling and image acquisition minimize these effects and produce images comparable to those obtained from conventionally prepared specimens. This streamlined workflow substantially reduces preparation time, lowers material requirements, and facilitates medium-throughput morphological analysis. The method provides a practical and reproducible approach for rapid developmental phenotyping and is readily adaptable to other plant species and other tissues.
The final morphology of a plant is determined by much earlier events that occur during development. Meristems are critical for the development of the aerial parts of plants, such as leaves and stems. The morphology of these meristems, which can vary significantly across genetic backgrounds, correlates with phenotypic traits in Zea mays (maize) that influence yield, such as leaf initiation rate and kernel row number. Traditional methods for morphological analysis of meristems, including histological sectioning and fine dissection, often require considerable technical expertise. To address these challenges, we present a detailed protocol employing methyl-salicylate for tissue clearing, combined with Differential Interference Contrast (DIC) microscopy to visualize maize shoot apical meristems. This method enables the intact visualization of vegetative meristems without the need for extensive dissection, thereby preserving the natural morphology of the meristem. Using this approach, the meristem is coarsely dissected, fixed, optionally stained, and cleared to enable visualization through leaf layers. By improving the accessibility and quality of meristem imaging, this protocol facilitates accurate quantification of meristem morphology, critical for genetic and developmental studies. Optimized primarily for maize, the outlined techniques can also be adapted for other plant species, providing a versatile tool in the field of plant developmental biology.
Advanced crop monitoring inside greenhouses is becoming one of the primary objectives of research centers. High-performance sensors, such as LiDAR or stereo cameras, have traditionally been employed for this purpose, though these often have a high cost. This work proposes a Visual-SLAM system using a monocular camera, which is significantly more cost-effective and specifically tailored for agricultural applications, such as mapping tomato crops in a greenhouse. Tests were carried out on a real tomato bunch, located in the Agroconnect experimental greenhouse. A ROS 2 Humble node was developed to run on the robot in order to capture images of these crops, which were then stored for offline processing. To generate a 3D mapped model for the crop in the greenhouse, the GLOMAP mapper, based on Structure-From-Motion, was integrated with the Hierarchical Localization toolbox. This initial mapping is a foundation for future, more advanced algorithms to analyze growth patterns, and optimize agricultural management. The system leverages a hierarchical localization paradigm based on a coarse-to-fine strategy: it first performs global retrieval to generate location hypotheses, then combines local features within the identified candidate regions. The results show a correct identification of the tomato cluster, correctly characterising the tomato that is occluded and inaccessible by classical vision technologies. The reconstructed 3D model was further validated against manual ground-truth measurements of fruit size, centroid position, and orientation, confirming the geometric accuracy of the proposed low-cost monocular pipeline.
Abstract Purpose Agrivoltaic vineyards show strong spatio-temporal variability in canopy shading, but field methods to quantify panel-induced shading at canopy scale remain limited. Shading is a key factor because it affects plant physiological and morphological traits, with potential consequences for yield and production quality. This study developed a near-surface time-lapse RGB imaging approach to derive temporally explicit shading metrics in an agrivoltaic vineyard of Vitis vinifera cv. Falanghina in Southern Italy. Methods Two representative vine positions beneath the photovoltaic structure were monitored: Agrivoltaic Shade (AVS), with greater exposure to panel-induced shading, and Agrivoltaic Light (AVL), with lower exposure. Image-based canopy shading percentage was calculated through a dedicated processing workflow and integrated with radiometric and physiological measurements, including continuous photosynthetically active radiation (PAR), canopy-level spectral photon flux measurements, photosynthetic photon flux density (PPFD), band-specific photon flux densities, red:far-red ratio (R:FR), stomatal conductance (gₛ), and leaf temperature. PAR measurements beneath the panels were compared with a full-sun control area. Results AVS showed significantly higher shading than AVL (76.14% vs 39.45%, p Conclusion The proposed workflow offers a low-cost, non-destructive tool to quantify shading dynamics and support site-specific assessment of crop microenvironments in agrivoltaic systems. The approach provides crop-relevant information for precision monitoring and management of spatially heterogeneous light conditions across different crop species. Impact The data provided in this manuscript enable the quantification of in-season photovoltaic-induced canopy shading dynamics in an agrivoltaic vineyard using proximal RGB time-lapse imaging and crop-level radiometric measurements. These metrics reflect the spatial and temporal variability of light availability within the vineyard and support site-specific assessment of crop microenvironments and precision management of agrivoltaic systems.
Polyploidy can be a critical factor for explaining plant trait variation, niche diversification, or speciation. However, inferring ploidy from silica-dried or historical samples using chromosome counts or flow cytometry is not possible, and scaling up ploidy estimation to population-level fresh contemporary samples can be challenging as well. Thus, we present a new method for estimating ploidy levels directly from sequencing data using machine learning; the Polyploid Population Genomics Tool Kit (PPGTK). The machine-learning approach is advantageous as it relaxes the assumptions of previous probabilistic methods and provides per-sample probabilities, allowing investigators to evaluate uncertainty in their system of interest.. We demonstrate performance and accuracy of the method on simulated and empirical data. Simulations showed above 99% accuracy, even for low coverage data, as long reads were mappable to the reference genome. For empirical analyses, we used target enrichment data from blueberry wild relatives (Vaccinium sect. Cyanococcus) and whole-genome data from sweetpotato wild relatives (Ipomoea ser. Batatas). Ploidy was recovered with 99% accuracy across 70 Vaccinium individuals and 97% across 82 Ipomoea individuals. Analysis of many individuals is fast and requires only a multisample VCF, which is presumably generated for the research anyway, and some samples of known ploidy for training the classifier. The approach implemented in PPGTK is promising for collections-based research as well, enabling ploidy classification of historical specimens based on present-day observations. The method is implemented in a new Python package as a single command that can run on a conventional laptop.
Reproduction assets foundThe paper's ploidy-classification method is implemented in the authors' public Python package PPGTK, with a specific release (v0.1.0-alpha) used for the manuscript's analyses. The empirical VCF/metadata datasets are promised on Dryad only 'upon acceptance' and thus are not yet actionable.Code · public11
VCFs and metadata needed to reproduce Vaccinium sect. Cyanococcus and Ipomoea ser.
372
Batatas analyses with PPGTK will be made available via Dryad upon acceptance. PPGTK is
373
available on GitHub, and release v0.1.0-alpha was the version used for analyses in this
374
manuscript (https://github.com/tileylab/PPGTK/releases/tag/v0.1.0-alpha). PPGTK currently has
375
other functions for calculating population genetic summary statistics, but the classify-ploidy
376
function implements the machine-learning method described in the manuscript.
377
378
.
CC-BY 4.0 International license
is made available under a
preprint (which was not certified by peer review) is the auOpen asset ↗tileylab/PPGTK · v0.1.0-alphapdf-raw-page:11 lines:1-24Plant phenotyping relevance matchOpenAlex · checked 15 Sept 2026
Photini V. Mylona · Lorenzo Barchi · Luciana Gaccione · Annalisa Cocozza · Pasquale Tripodi · Ilias Avdikos
Horticultural crops, particularly Solanaceae and Cucurbitaceae, represent a major component of global vegetable production and are increasingly exposed to climate variability and environmental constraints. Landraces and crop wild relatives constitute essential reservoirs of adaptive genetic diversity; however, their effective utilization in breeding programs remains limited by fragmented characterization, incomplete passport information, and reliance on labor-intensive morphological descriptors. These limitations hinder the systematic exploitation of conserved germplasm and restrict its integration into modern predictive breeding frameworks. Recent advances in high-throughput phenotyping, genomics, and multi-omics technologies have created new opportunities to bridge the gap between genotype and phenotype. Next-generation phenomics enables non-destructive, high-resolution quantification of plant physiological and structural traits across environments, while genomic approaches, including whole-genome resequencing, support comprehensive assessment of genetic diversity. Pangenome frameworks further extend this resolution by capturing core and variable genomic fractions, collectively defining the species variome and enabling improved identification of structural and allelic variants associated with adaptive traits. The integration of phenomic and genomic datasets through multi-omics approaches enhances the functional interpretation of trait-associated variation and strengthens the predictive capacity of breeding strategies. In this context, the genome as a functional passport constitutes a unified reference layer linking germplasm identity with genomic, phenotypic, and functional trait information, thereby enabling more systematic germplasm characterization, reduced redundancy, and improved identification of elite parental material. This mini-review highlights how the integration of phenomics, pangenomics, and multi-omics enables the transition from descriptive germplasm cataloguing toward more systematic, data-driven, and predictive breeding systems for the development of climate-resilient horticultural crops.
Abstract Background Root architecture determines the capacity of crops for spatial exploration under stress conditions; however, existing studies on salt tolerance screening have mostly been confined to single traits such as root length or biomass, overlooking the overall spatial configuration of the root system and its intrinsic linkage with aboveground physiological functions. On this basis, the present study aimed to determine whether root convex hull area can characterize root–shoot synergistic adaptability under salt stress, and whether this synergy involves a physiological mechanism of resource conservation through cortical tissue remodeling. Methods Using a paper-based root phenotyping platform, we screened 28 spring wheat varieties originating from the arid regions of northwest China under 200 mM NaCl stress, and compared the differences between large-convex hull area and small-convex hull area varieties in root architecture, root cortical anatomy, stomatal traits, leaf water status, photosystem II efficiency, canopy temperature, and transpiration rate under salt stress. Results The results showed that large-convex hull area varieties maintained total root length, maximum depth, and convex hull area under salt stress, whereas small-convex hull area varieties exhibited significant reductions in all these parameters. Meanwhile, compared with small-convex hull area varieties, large-convex hull area varieties possessed greater cortical lacunar tissue area and cortex/stele ratio, as well as higher stomatal density, leaf relative water content, F v/ F m, and transpiration rate, but lower canopy temperature and smaller stomatal aperture. Convex hull area was positively correlated with leaf water status, photochemical efficiency, cortical lacunar area, and stomatal density, while negatively correlated with canopy temperature and stomatal aperture, indicating that root spatial maintenance, moderate cortical senescence, and stomatal regulation together constitute a functionally coordinated response module under salt stress. Conclusion In summary, convex hull area is not merely a descriptive indicator of root morphology, but rather a functional trait that reflects the synergistic integration of belowground exploration capacity and aboveground physiological resilience. This study proposes that convex hull area can serve as a candidate high-throughput phenotypic indicator for salt tolerance screening in wheat at the seedling stage; nevertheless, its predictive capacity for field yield performance still requires further validation under soil conditions, across the full growth cycle, and under interactions with multiple environmental factors.
Background and aims Recognizing lineages is a central challenge in plant systematics, making it essential to explore multiple analytical tools. In this context, this study investigates how frond shape can assist in discriminating against lineages within the Scaly clade of Microgramma (Polypodiaceae), and tests whether the integration of multiple lines of evidence enables a more consistent recognition of lineages than exclusively macromorphological approaches. Methods We analyzed 271 specimens representing eight species, using Elliptical Fourier Analysis (EFA) to quantify frond shape, followed by multivariate statistical tests (PCA, MANOVA, LDA). Evolutionary relationships between spectral and morphometric data were assessed through phylogenetic generalized least squares (PGLS) regressions and phylogenetic partial least squares (Phylo-PLS) analyses. Results Dimorphic species exhibited higher discrimination capacity (average accuracy of 80-83%). Fertile and combined fronds yielded the highest accuracy values. Morphologically similar species, such as M. reptans and M. tobagensis, showed considerable overlap, whereas M. percussa achieved the best performance (average accuracy of 80%). Morphometric-spectral covariation yielded R2 = 0.72 (P = 0.003) when sterile and fertile fronds were combined, although PGLS analyses detected no significant phylogenetic signal. The association differed between frond types and is interpreted here as exploratory evidence of morpho-spectral covariation. Conclusions Outline morphometry combined with infrared spectroscopy within a phylogenetic framework improves lineage discrimination, although overlap zones persist, reflecting complex evolutionary processes. Our study highlights the potential of integrative systematics to elucidate species boundaries in groups with high morphological disparity, as well as the need for broad sampling and multi-evidence approaches in future systematic reviews.
Balem JM, Tan C, Dias NCF, Arnold M, Tran S, Severns PM, Teixeira PJPL, Li C, Yang L.
ArabidopsisThermalLeafTissueGrowth / time-series analysisStress response / tolerancePlant / canopy temperature
Repairing damaged tissues is essential for the survival of all organisms. In plants, tissue injury rapidly triggers defense and repair programs. However, the molecular mechanisms linking early injury cues to the later stage of wound repair remain unclear. Here, we show that wounding of Arabidopsis leaves induces localized low temperature at the injury site, likely caused by evaporative cooling, which is accompanied by an activation of cold-responsive genes. Using thermal imaging combined with computer vision and deep learning, we developed a workflow to monitor the dynamics of wound healing in a quantitative, non-invasive, and real-time manner. Mechanistically, we show that C-repeat Binding Factor (CBF) transcription factors are required for the activation of the injury-associated cold response and downstream salicylic acid (SA) signaling. Our findings suggest that the CBF-SA pathway acts coordinately to promote lignin and callose deposition, thereby facilitating wound repair. Together, these findings reveal a link between a wound-induced biophysical cue and the tissue repair program.
Abstract Context: Accurate leaf area index (LAI) estimation is essential for UAV-based winter wheat growth monitoring, but optical saturation effects under dense canopies remain a major limitation. Aims: This study aims to develop a lightweight dual-stream network (HFI-Net) that improves LAI estimation under high-LAI conditions where saturation effects commonly occur in conventional vegetation indices for efficient UAV-based LAI mapping. Methods: A field dataset containing 637 paired UAV image patches and ground LAI measurements was collected from 21 winter wheat cultivars across seven phenological stages. HFI-Net was proposed, integrating RGB texture and vegetation-index (VI) features derived from multispectral imagery through attention-guided star-shaped multiplicative feature interactions, and five-fold cross-validation with data augmentation was applied. Key Results: HFI-Net achieved a coefficient of determination (\((R^2)\)) of 0.9023 and an RMSE of 0.4822 on the test set. Compared with ResNet50, the proposed model reduced parameters by 98.3% to only 0.40 M, while providing improved prediction performance under saturation-prone high-LAI conditions (\((>4.0)\)). Conclusion: HFI-Net enables reliable winter wheat growth monitoring throughout the growing season with reduced performance degradation under high-LAI conditions and low computational cost. Implications and Impacts: These results indicate the potential of HFI-Net for efficient UAV-based LAI mapping and precision agriculture applications.
Abstract Image-based plant-disease classifiers often achieve high accuracy under controlled conditions, but their reliability can deteriorate in real-world field environments. This article investigates crossdataset strawberry leaf-health classification using controlled PlantVillage images and field images collected through the SmartBerry project in Nigeria. Four deep-learning architectures were evaluated under within-domain, controlled-to-field and pooled-training conditions using healthy-leaf and darkspot/ scorch symptoms. Although all models achieved near-ceiling performance on PlantVillage, direct transfer to SmartBerry reduced macro F1-score to between 0.418 and 0.824, demonstrating a substantial gap between controlled-dataset performance and field reliability. To reduce the amount of locally labelled field data required to address this gap, this paper proposes SmartBerry Domain Adaptation (SmartBerry-DA). This label-efficient adaptation approach combines a model trained on controlled images with a small set of labelled SmartBerry images and additional unlabelled field images. Using MobileViTv2-0.5, SmartBerry-DA achieved a macro F1-score of 0.940 ± 0.015 with only five labelled SmartBerry images per class and 0.988 ± 0.012 with ten, compared with 0.994±0.008 when the full labelled target dataset was used. This shows that strong field performance can be recovered with substantially reduced local annotation. For lightweight offline deployment, 16- bit floating-point representation reduced model size from 4.37 MB to 2.30 MB without changing predictive performance under the evaluated conditions. The findings demonstrate the importance of locally acquired field data and label-efficient adaptation for transferring agricultural AI from benchmark datasets to practical farming environments, while providing a pathway towards lightweight strawberry disease recognition in settings where extensive annotation and continuous connectivity may not be feasible.
Reproduction assets foundThe paper's own SmartBerry field strawberry image dataset is explicitly stated to be publicly available on Zenodo with a Kaggle mirror. PlantVillage is a cited third-party benchmark, and no author analysis code or trained model checkpoints are stated as available.Dataset · publicip with the SmartBerry partners.
Ethics statement
NA
Data availability
The PlantVillage dataset is publicly available from its
cited source. The SmartBerry dataset is publicly avail-
able through Zenodo, with a Kaggle mirror provided for
convenient access.SmartBerry dataset: Zenodo: https://
zenodo.org/records/22056372; Kaggle: https://www.kaggle.com/datasets/isiborihianle/smartberry-strawberry-field-image-dataset
Funding
This research was supported by the SmartBerry Project
(Project Number: 10071867) and the Department of Com-
puter Science, Nottingham Trent University.
Acknowledgements
The authors thank the Ashley Strawberry Farm staff who
supported image collection, annotation and field eOpen asset ↗Kaggle · isiborihianle/smartberry-strawberry-field-image-datasetpdf-raw-page:17 lines:1-89Plant phenotyping relevance matchCrossref · checked 15 Sept 2026
Published9 Sept 2026Journal of Applied Science and Technology Trends
This study proposes an integrated hybrid pipeline approach which is the combination of an optimized Inception-v4 Convolutional Neural Network (CNN), YOLOv8 (You Only Look Once version 8) and extreme gradient boosting (XG Boost) models to detect and classify Tomato plant diseases effectively and accurately. The proposed methodology is based on the fine-grained localization ability of the YOLOv8 model to accurately localize the affected area of the leaf, multi-scale deep feature extraction ability of the optimized Inceptionv4 CNN, and XG Boost to reduce dimensions and optimize features. The hybrid model combines the Inception-v4 CNN, YOLOv8, and XG Boost models with an average CPU processing time of about 200 milliseconds per image, the model performs better in terms of computational efficiency, robustness, and faster prediction. Experimental tests on Plant Village show that the model outperforms the state of the art in 10 disease categories. Specifically, the integrated hybrid system achieved high precision, recall, F1-score, mean Average Precision (m AP), and its training accuracy was more than 0.98 (up to 0.9969), and the testing accuracy was more than 0.96 (up to 0.9695). The hybrid framework shows good reliability even for diseases that are visually similar, such as bacterial spot, early blight, late blight, mosaic virus and yellow leaf curl virus. The optimization of batch size to 32, along with the tuning of the learning rate further improved the stability of training, convergence speed and the generalization of the overall model. The proposed system holds promise for real-time, scalable, and sustainable precision agriculture, aiming for early disease detection and yield protection. Future work includes incorporating Internet of Things (IoT) edge devices, in field environmental monitoring, multimodal data integration and applying Explainable Artificial Intelligence (EAI) techniques to enhance model interpretability for end-users.
Kelsey M. Reed · Abigail A. Masri · Andrew J. Hanrahan · Emery L. Ng · Mao Li · Bastiaan O. R. Bargmann
ArabidopsisLaboratory / benchtopMicroscopyCell / cellular structureGrowth / time-series analysisTrackingVisualization / data managementGrowth / development / phenology
Plant developmental biology lacks cell-based experimental systems comparable to the organoids and live-imaging platforms that have transformed mechanistic discovery in animal research. To address this gap, we present a robust, trackable protoplast regeneration platform in Arabidopsis thaliana that enables high-resolution, time-resolved analysis from the single-cell stage through microcolony formation and early regenerative development. Protoplasts are embedded in thin alginate matrices containing fluorescent fiducial beads, maintaining physical separation and allowing repeated return to the same cells over days to weeks. This design supports long-term imaging using epifluorescence, confocal, and lattice light sheet microscopy, enabling visualization of cell-cycle re-entry, asymmetric division, organelle dynamics, dedifferentiation, redifferentiation, and regenerative competence. Fluorescent reporters for nuclei, membranes, microtubules, Golgi, and hormone signaling further permit observation of subcellular organization and signaling heterogeneity during early reprogramming. Together, this platform provides an accessible, scalable system for studying plant cellular plasticity at single-cell resolution and offers a foundation for plant organoid-like models. By enabling high-resolution study of regeneration from single cells, this method expands the experimental toolkit and supports efforts to overcome species- and genotype-dependent barriers to transformation and plant biotechnology.
Abstract Accurate estimation of maize leaf nitrogen content is important for improving nitrogen-use efficiency and supporting precision crop management. However, leaf-level hyperspectral modeling is challenged by high spectral redundancy and heterogeneous spectral responses among local leaf regions. This study proposes a Hyperspectral–Region Aggregation Network (HSRAN) for maize leaf nitrogen content estimation from region-level hyperspectral spectra. HSRAN consists of a Spectral Adaptive Recalibration Encoder (SARE) and a Context-Aware Gated Aggregation Module (CAGM). SARE performs band-wise residual recalibration and extracts regional spectral representations, whereas CAGM models contextual dependencies among regional features and performs gated attention-based aggregation for leaf-level prediction. Field experiments were conducted in 2024 and 2025 at the jointing, silking, and maturity stages. HSRAN was evaluated against PLSR, RF, XGBoost, SVR, 1D-CNN, MLP, and Transformer1D models. Across the stage-specific and pooled datasets, HSRAN achieved the highest R² and the lowest RMSE while maintaining competitive MAE values. On the pooled full-growth-period dataset, HSRAN achieved an R² of 0.84, an RMSE of 3.63 g kg⁻¹, and an MAE of 2.59 g kg⁻¹. At the jointing, silking, and maturity stages, the corresponding R² values were 0.56, 0.76, and 0.72, respectively. Ablation experiments indicated that integrating SARE and CAGM improved R² from 0.80 to 0.84. To interpret regional contributions, the learned attention weights were mapped back to the original leaf coordinates recorded during regional sampling. Regions near leaf veins, tips, and margins often received relatively higher attention weights, suggesting that their local spectra provided informative cues for model prediction. These findings indicate that spectral–regional joint modeling can improve leaf-level hyperspectral estimation of maize nitrogen content. HSRAN provides a practical framework for non-destructive nitrogen assessment in maize.
Doolan O, Peirats-Llobet M, Lewsey MG, Aberdein N, Bricklebank N.
BarleySeed / grain
Quantitative imaging of plant tissues by laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) is hindered by the lack of matrix-matched calibration standards. Established approaches, such as gelatine or homogenised tissue blocks, do not replicate plant matrices accurately. Here, we introduce a nano-dispenser-based calibration strategy that deposits nanolitre volumes of elemental standards directly onto paraffin-embedded grain sections, exploiting the low endogenous metal content of the endosperm to generate in situ calibration curves. Calibration performance for Mg, Mn, Cu, Zn, and Mo was assessed using LA-ICP-MS imaging and Iolite 4 data processing. The method demonstrated excellent linearity (R 2 > 0.98), reproducibility across multiple grains, and sub-ppm limits of detection. Comparative analysis with in-house homogenised blocks and NIST wheat reference material confirmed superior accuracy and reproducibility of the nano-dispenser approach. As a proof-of-concept, we have applied the method for the quantitative imaging of metals in a whole barley grain section, and the results show excellent agreement with published data obtained by conventional liquid-mode ICP MS. The technique reported here provides a robust, scalable, solution for quantitative metallomics studies of plant tissues, enabling improved assessment of nutrient distribution and supporting the development of standardised protocols for LA-ICP-MS imaging.
Abstract Purpose The study aimed to develop a hybrid Physics-Informed Neural Network (PINN) architecture to forecast the daily strawberry canopy volume using weather data and image-derived inputs. Green pixel counts were used as a low-cost proxy for canopy volume and tracked canopy growth in two strawberry cultivars under real-world field conditions. The PINN architecture was chosen because it directly embeds known growth constraints into the learning process, enabling reliable predictions even with a few observations per day. Methods Two commercially important strawberry ( Fragaria × ananassa Duch.) cultivars, ‘Florida Brilliance’ and ‘FL 16.30–128’ (marketed as Florida Medallion™) were used in this study. Strawberry plant images were acquired at 15-minute intervals over 38 days and were used to train the PINN architecture along with 13 weather data points and accumulated Growing Degree Days (GDD). The PINN was composed of a Long Short-Term Memory (LSTM) residual encoder and a trainable logistic growth curve driven by GDD. Biologically implausible canopy decline during GDD summation increases was penalized by the loss function. Results The Brilliance plant model yielded a Mean Absolute Error (MAE) of 6,996 pixels of canopy size and a Mean Absolute Percentage Error (MAPE) of 4.53%, while the Medallion plant model produced an MAE of 9,440 pixels and an MAPE of 16.4%. Both models outperformed a naive persistence baseline, which produced MAE values of 9,416 and 9,580 pixels for Brilliance and Medallion, respectively, and a physics-only logistic baseline, which yielded an MAE of 10,260 and 10,880 pixels for Brilliance and Medallion, respectively. Conclusion Combining a straightforward growth physics prior with a short-sequence LSTM significantly improved canopy forecasting accuracy, even with about a month of training instances. The system could be used to investigate strawberry growth during the vegetative and early fruiting periods to improve management and increase profit.
Jialiang Zheng · Qingmin Pan · Yixue Zhang · Chuandong Guo · Hanping Mao · Xiaodong Zhang
Physiological trait estimation
Plant wearable sensors have emerged as a transformative technology for precision agriculture and plant phenotyping, enabling in situ, real-time, and continuous acquisition of physiological signals from plant surfaces or internal tissues. However, existing reviews have organized the literature by monitoring targets, sensing functions, or material platforms, without systematically comparing technologies from the fundamental dimension of the degree of intervention imposed on plants. Drawing on representative studies identified through a structured literature search, this review establishes a three-tier classification framework—invasive, minimally invasive, and non-invasive—and conducts a head-to-head comparison across six dimensions: signal characteristics, plant disturbance, long-term stability, manufacturing complexity, field deployability, and biosafety. The results reveal that invasive sensors (nanobionic probes, implantable microelectrodes, and organic electrochemical transistors) achieve nM–pM detection limits, yet wound responses generally limit their effective monitoring duration to the order of days; non-invasive sensors (flexible patches, strain sensors, and multimodal platforms) support weeks-to-months of continuous monitoring and are amenable to scaled deployment, but the indirectness of surface signals confines detection limits to the μM level; minimally invasive technologies (microneedle arrays and ultra-thin microelectrodes) offer a compromise between the two extremes. On this basis, a decision framework based on three-layer selection is proposed to guide technology selection across laboratory research, field deployment, and controlled environment agriculture. Future efforts should focus on standardized performance evaluation protocols, biodegradable self-powered systems, and the integration of invasive–non-invasive hybrid sensing networks with plant digital twins.
Abstract Tomato leaf diseases (TLDs) such as Early Blight (EB), Late Blight (LB), and Leaf Mold (LM) have a negative impact on crop yield and quality, result-ing in economic losses in agriculture. Traditional diagnosis methods depend on the visual recognition of a disease, which are time-consuming, subjective and may be difficult to reach in rural settings. Keeping these drawbacks in mind, the present work introduces a Transfer Learning model for tomato leaf disease classification based on EfficientNetB3 network. A pretrained Ef-ficientNetB3 network, trained on ImageNet, is used to obtain discriminative features like lesion boundaries, discoloration, fungal textures, and infection spots. Images are cropped to 224 × 224 pixels and split into training, validation and test set. The proposed architecture employs Global Average Pooling, a dense layer with ReLU activation, drop out regularization and Softmax classifier for classification of tomato leaf images into four classes: Early Blight, Late Blight, Leaf Mold and Healthy. Experimental results show the high classification accuracy and low training, validation and test-ing losses. Inter-class misclassifications of the confusion matrix show very few, which supports good generalization. Moreover, Grad-CAM visualiza-tions can generate interpretable heat maps that point to the regions of the image affected by the disease, and multi-class ROC analysis gives high AUC values, which means that it has excellent class separability. The proposed framework provides a precise, reliable, and interpretable approach for auto-mated tomato leaf disease diagnosis, contributing to precision agriculture by assisting in early detection and prompt management of tomato diseases.
Reproduction assets foundThe paper's tomato leaf disease classification uses a publicly available Kaggle dataset (PlantVillage-derived tomato leaf images, 4000 images across 4 classes), explicitly declared in the Data Availability statement with a public URL. No author code, trained models, or other paper-specific assets are disclosed.Dataset · publicitted in accordance with the Journal policies.
Permission to use third-party material
Images or figures are never published previously and did not take from any
internet resources.
Data Availability
The dataset used in this study is publicly available from the PlantVillage
tomato leaf disease dataset on Kaggle’s following link:
https://www.kaggle.com/datasets/kaustubhb999/tomatoleaf
Acknowledgements
The authors would like to thank SR University, Warangal, Telangana,
INDIA and Sharda University, Greater Noida, Uttar Pradesh, INDIA for
providing research facilities and computational resources to carry out this
work.Open asset ↗Kaggle · kaustubhb999/tomatoleafpdf-raw-page:20 lines:1-18Plant phenotyping relevance matchOpenAlex · checked 15 Sept 2026
Broad adoption of chlorophyll fluorescence kinetics in crop phenotyping remains limited by inconsistent parameter definitions and insufficient physiological validation. This study employed a longitudinal phenomic approach to evaluate the diagnostic value of rapid chlorophyll a fluorescence kinetics across diverse wheat genotypes throughout a spring growing season. By combining JIP-test parameters with detailed growth analysis, we built a statistical framework quantifying the predictive power and unexplained variance of biophysical parameters derived from the JIP-test, with an emphasis on performance-estimating indices. Results show that conventional parameters, such as maximum quantum yield of primary photochemistry, remain stable during vegetative growth and are highly sensitive to terminal senescence, yet display considerable noise under transient environmental fluctuations. Conversely, integrative indicators, particularly the total performance index, revealed clear seasonal trends and strong correlations with growth dynamics. By grouping JIP-test parameters functionally, this study offers a rational framework for selecting biophysical indices suited to specific breeding or research goals.
Volatile aroma compounds (VACs) are key semiochemicals induced and released by plants under biotic and abiotic stresses. Therefore, it is of great significance for detecting VACs to evaluate the growth environment information and physiological markers of plants. However, real-time and on-site monitoring of VACs at the current stage remains a challenge due to their complex species and trace concentrations. Here, we report an ultrasensitive VAC sensor based on Ru-nanocluster-anchored ZnO nanodisks, for which the corresponding limit of detection for VACs can reach as low as 10 ppb. By constructing temperature-modulated 4 × 4 VAC sensor arrays and integrating artificial intelligence algorithms, we demonstrate simultaneous identification of 10 VAC species and concentrations with 98% accuracy and high fidelity. Furthermore, a portable biomimetic olfactory e-nose based on temperature-modulated sensor arrays was developed to realize the on-site detection of VACs released from tea plants induced by Ectropis obliqua. This work not only provides a design strategy for ultrasensitive VAC sensors but also establishes a scalable platform for real-time plant health monitoring and early warning of biotic and abiotic stresses.
Abstract Jute is one of the most important cash crops in Bangladesh and plays a vital role in the country's economy. However, the yield and quality of jute are often affected by various plant diseases, which are difficult to detect manually at an early stage. Early identification is essential to prevent large-scale damage and ensure sustainable production. This study proposes an automated jute disease detection system using image processing and deep learning techniques. Several models, including a Custom Convolutional Neural Network (CNN), VGG16, DenseNet121, and a Hybrid EfficientNetB7 architecture, were implemented and compared. The proposed Hybrid EfficientNetB7 model achieved the highest classification accuracy of 99.7%, outperforming the other models. The system effectively distinguishes between healthy and diseased jute leaves and stems, providing a reliable tool for early disease detection. This research demonstrates that deep learning-based ensemble and transfer learning methods can significantly improve the accuracy and reliability of plant disease detection systems. The findings can contribute to the development of intelligent agricultural tools, empowering farmers with faster and more accurate disease diagnosis for better crop management and yield improvement.
Abstract DenseNet121 is effective for rice disease recognition, but its dense feature reuse keeps many connections active during inference. We evaluated whether learnable layer gates could improve paddy disease classification and expose removable DenseNet connections. Using the Kaggle Paddy Disease Classification dataset, we trained an ImageNet-pretrained DenseNet121 baseline and an Adaptive Gated DenseNet121 on 8,325 training images and validated on 2,082 images across ten classes. The baseline achieved 96.78% accuracy and 96.64% macro F1-score. The adaptive model achieved 97.21% accuracy and 96.89% macro F1-score, with 97.16% macro precision and 96.65% macro recall. Pruning at threshold 0.825 reduced parameters from 6.96 million to 6.84 million while retaining 96.39% F1-score, but slightly higher thresholds caused severe collapse. Adaptive gating therefore improved classification and enabled limited pruning, but did not yet provide robust mobile compression.
Reproduction assets foundThe paper uses the public Kaggle Paddy Disease Classification image dataset for all phenotyping measurements and states that the authors' PyTorch implementation, Kaggle notebooks, and report-generation scripts are publicly available in a GitHub repository. Both are paper-specific, public, and actionable.Dataset · publicarticipants, human tissue, live vertebrates or higher invertebrates.
Use of artificial intelligence assistance
A large language model was used to assist with code organization and formatting.
Data availability
The dataset presented and used in this study are the publicly available Paddy Disease Classification dataset on Kaggle: https://www.kaggle.com/competitions/paddy-disease-classification/data.
Code availability
The PyTorch implementation, Kaggle notebooks and report-generation scripts are available in the project workspace on Github [20].
Declarations
Acknowledgements
The authors acknowledge the Kaggle Paddy Disease Classification competition and the Paddy Doctor dataset contributors forOpen asset ↗lines:177-215Code · publicThe PyTorch implementation, Kaggle notebooks and report-generation scripts are available in the project workspace on Github [20].Open asset ↗lines:177-215Plant phenotyping relevance matchOpenAlex · arXiv · checked 15 Sept 2026
Daniel Rosendo · Renan Souza · Kelsey Carter · John Lagergren · Frédéric Suter · Shelaine Curd · David Weston · Rafael Ferreira da Silva
Laboratory / benchtop
Autonomous, cross-facility science requires capabilities that no individual project should have to build for itself: managed execution for long-lived services, versioned distribution of models to remote compute systems, governed access to large language models, a shared substrate for experimental data, and end-to-end provenance. The U.S. Department of Energy Genesis Mission platform, delivered through the American Science Cloud, provides these as reusable services. This paper reports how the Genesis platform enables cross-facility experiments and accelerates scientific discovery. We explore the plant phenotyping workflow of the Orchestrated Platform for Autonomous Laboratories as the exemplar: it couples Oak Ridge National Laboratory's Advanced Plant Phenotyping Laboratory with the Frontier supercomputer. In a 40-day nickel-treatment campaign, the resulting workflow replaced roughly twelve hours of manual analysis with interactive queries returning in seconds to minutes.
Abstract Background and Aims Drought reduces wheat yields, yet field-scale quantification of root water uptake (RWU) remains challenging because below-ground processes are difficult to monitor. This study developed a non-invasive hydrogeophysical framework integrating Electrical Resistivity Tomography (ERT), TDR-based soil monitoring, and depth-aware Random Forest calibration to quantify depth-resolved RWU and evaluate genotype-specific water-use strategies under terminal drought. Methods Time-lapse ERT (44 surveys, ≥ 3 week⁻ 1 ) was combined with TDR sensor measurements of soil water content (n = 278 paired ρ–θ observations) to convert resistivity measurements into depth-resolved RWU estimates across 0.1–1.0 m depth. Five petrophysical models were evaluated using date-grouped fivefold cross-validation, with the depth-aware Random Forest performing best. Three wheat genotypes with contrasting root architectures were monitored under terminal drought (142 mm available water). ERT-derived RWU were analysed alongside stomatal conductance, chlorophyll fluorescence, and grain yield. Results ERT resolved RWU strategies among genotypes. WM-203 exhibited aggressive, coordinated multi-layer water extraction across the soil profile (r = 0.80–0.98), whereas WM-140 showed a delayed uptake strategy characterized by early deep-layer dominance followed by mid- and deep-profile engagement, and IPLR-760 displayed inconsistent uptake with mid-profile hydraulic decoupling. Genotypic RWU rankings were consistent with stomatal conductance and grain yield, spanning from 7.0 t ha⁻ 1 in WM-203 to 1.5 t ha⁻ 1 in IPLR-760 despite comparable total water extraction. Conclusion ERT-based quantification of RWU provides a robust, non-invasive approach for resolving genotype-specific water-use strategies under field conditions. The framework enables characterization of water-use coordination patterns and offers a tool for phenotyping drought-resilient wheat genotypes.
Reproduction assets foundThe paper's Data availability statement explicitly states that the code and supporting data for this ERT-based root water uptake study are publicly available on the authors' GitHub repository, which is listed in allowed_urls. This qualifies as a paper-specific public code/data asset for the phenotyping analysis.Code · publicsity of Jerusalem. This research was supported by the Chief
Scientist of the Israeli Ministry of Agriculture and Food Secu-
rity (grant no. 12–01-0056) and the Israeli Council for Higher
Education (Project: Future Crops for Carbon Farming).
Data availability The code and supporting data for this study
are publicly available at:
https://github.com/emmaiyke/ERT_RWU_Wheat_Project
Additional datasets are available from the corresponding
author upon reasonable request.
Declarations
Competing interests The authors declare that they have no
known competing financial interests or personal relationships
that could have appeared to influence the work reported in this
paper.
Open Access This article isOpen asset ↗ERT_RWU_Wheat_Project · emmaiyke/ERT_RWU_Wheat_Projectpdf-raw-page:22 lines:1-95Plant phenotyping relevance matchEurope PMC · checked 15 Sept 2026
Cadmium contamination severely affects rice growth, yield, and quality, making early stress monitoring essential for agricultural management and food safety. However, traditional detection methods are cumbersome and time-consuming, limiting their applicability to early stress diagnosis. This study developed a rapid and accurate approach for discriminating cadmium stress levels in rice. Arginine-modified flower-like silver nanoparticles (Ag NPs-Arg) were synthesized to enhance Raman signals associated with three stress-response indicators: salicylic acid (SA), malondialdehyde (MDA), and peroxidase (POD) activity. Quantitative prediction models for these physiological indicators and a stress-level discrimination model were established. Among the evaluated models, the CNN-Transformer model achieved the best predictive performance, with Rp 2 values of 0.889, 0.832, and 0.802 for SA, MDA, and POD activity, respectively. An objective weighting method was used to integrate the three biochemical reference indicators, providing a multi-indicator physiological basis for comprehensive stress assessment. The resulting stress-level assessment model achieved an accuracy of 95.83%, demonstrating its ability to capture cadmium-induced physiological changes and assess stress levels in rice.
Plant diseases, being a subject of interdisciplinary research, significantly reduce crop yield, quality, and economic returns, while the misidentification of pathogens often leads to ineffective treatments and may harm beneficial organisms and ecosystems. This work develops an approach for robust visual classification of plant diseases under limited and heterogeneous data based on multi-scale fractal texture descriptors integrated into a convolutional neural network. The proposed method employs wavelet transform modulus maxima to extract two complementary fractal characteristics, local fractal dimension and singularity spectrum width, from leaf images at several spatial scales. These descriptors form multi-channel fractal maps fed into a fractal attention module (FAM) inserted after the third stage of a ResNet-50 architecture. The FAM learns to emphasize spatial regions where fractal properties are most discriminative, while a parallel branch encodes global fractal statistics into an auxiliary vector combined with backbone features at the final classification layer. Experiments are conducted on a large heterogeneous collection of 11 public plant disease datasets under 5-shot, 50-shot, and full-scale training regimes. The fractal-augmented model raises classification accuracy from 57.06% to 67.73% on 5 shots and from 80.81% to 86.11% on 50 shots, red outperforming the plain ResNet-50 in these settings, converges within 1–2 epochs versus 25–40, and shows markedly better resilience to color distortions, random occlusions, and grayscale conversion in most cases. The generated attention maps provide spatially explicit explanations of the model’s decisions, increasing transparency for practical use. The proposed approach demonstrates that fractal analysis, embedded as a modulating signal inside a deep network, can serve as an efficient and interpretable inductive bias, which is particularly valuable under data scarcity and noisy agricultural imagery.
Field / plotFlowerClassificationObject detectionGrowth / development / phenology
Abstract Dioecious crops face significant pollination challenges due to the asynchrony in flowering between male and female plants. This asynchrony varies spatially across orchards, requiring targeted interventions in zones where synchrony is lacking. Assisted pollination addresses this deficiency, albeit at a substantial operational cost that could be optimised through spatial phenological mapping. Manual assessment proves economically infeasible at commercial scales, while existing computer vision systems are unable to classify phenological stages and integrate geospatial information. This study presents the Mobile Phenological Mapping (MPM) Framework, which integrates GNSS-synchronised smartphone video with automated phenological detection to generate plant-level phenological distribution maps validated in commercial kiwifruit (Actinidia chinensis) orchards. This modular framework comprises (i) training and validation data acquisition, (ii) model optimisation, (iii) operational pipeline, and (iv) performance evaluation. MPM employs hierarchical deep learning across three stages: structure detection, gender classification, and phenological stage classification. Video frames are georeferenced through timestamp matching with GNSS metadata, enabling spatial phenological mapping. Operational validation across four commercial orchard zones demonstrated mean absolute percentage errors of 17.2% for structure detection and 20.1% for gender classification. The framework reduces monitoring time from 113 to 1.6 hours per hectare, decreasing labour costs from €2 060 (113 hours × 18.20 € per hour) to €29 (1.6 hours × 18.20 € per hour) per hectare based on the Portuguese hourly labour cost for minimum wage workers. When integrated with routine orchard operations, video acquisition incurs negligible additional cost. MPM provides growers with precision phenological maps for targeted pollination interventions. While validated in a kiwifruit orchard, the modular architecture can be adapted to other crops by replacing the training data.
This study systematically evaluated the contribution of UAV LiDAR structural features such as crop height (CH) and multi-layer gap fraction (GF) and the amplitude of the returning signal represented by normalized intensity (INT), together with multispectral (MS) and thermal infrared (TIR) observations for aboveground biomass (AGB) estimation in winter wheat using a common artificial neural network (ANN) framework. Among the evaluated single sensor approaches, LiDAR features consistently provided the strongest performance, demonstrating the complementary value of crop height, vertically distributed canopy density, and normalized LiDAR intensity for characterizing canopy structure and within-canopy variability. Multi-layer GF improved AGB estimation relative to conventional ground-based GF approaches, highlighting the importance of incorporating the vertical distribution of canopy density. Multi-sensor fusion produced only modest additional improvements, indicating limited benefits relative to the increased acquisition and processing requirements. Temporal analysis showed that structural LiDAR features were most informative during early crop development, whereas normalized intensity, spectral reflectance, and thermal observations became increasingly valuable during canopy maturation and senescence. Comparisons with destructively measured plant area index (PAI), leaf area index (LAI), green leaf area index (GLAI), and green fraction of LAI further demonstrated that normalized LiDAR intensity (903 nm) was more closely associated with green canopy components than purely structural LiDAR metrics. Overall, the results demonstrate that fully exploiting both the structural and spectral information contained within LiDAR observations can substantially improve UAV-based biomass estimation, while multispectral and thermal observations provide complementary information whose contribution varies with crop development and monitoring objectives.
Accurate litchi counting from whole-tree images is essential for yield estimation, orchard management, and plant phenotyping, but remains challenging in real orchards because fruits occur in dense, heavily occluded clusters and vary markedly in scale, illumination, and appearance across ripening stages, particularly when green fruits resemble surrounding foliage. Existing methods have shown promise, but their robustness in complex orchard environments remains limited. To address these challenges, we propose FG-LCNet, a two-stage foreground-guided litchi counting framework. In the first stage, an enhanced fruit-cluster detector improves the localization of small and ambiguous clusters under complex canopy backgrounds. In the second stage, the detected foreground regions are fed into a density-regression network with hybrid attention, while a consistency-based training strategy is introduced to improve robustness to appearance and illumination variations. To support this study, a large-scale litchi counting dataset was established, consisting of 1,126 whole-tree images collected from five orchards and spanning three ripening stages, with approximately 120,000 fruit-level dot annotations and more than 20,000 cluster-level bounding boxes. FG-LCNet achieved the best overall counting performance, with an MAE of 7.44 and an RMSE of 11.01. It showed clear advantages in high-density fruit-cluster scenarios and cross-orchard validation, while maintaining competitive results across orchard-region and maturity-stage subsets. The framework further retained inference efficiency suitable for practical deployment. These results indicate that FG-LCNet provides an effective solution for robust litchi counting and offers potential for other clustered fruit-counting tasks.
Reproduction assets foundThe paper's implementation code is explicitly stated as publicly available at the authors' GitHub repository (FG-LCNet). The litchi counting dataset (1,126 whole-tree images with ~120,000 dot annotations and 20,000+ bounding boxes) is not yet fully public: a ~100-image annotated subset is promised upon acceptance, and,Code · publicdustry Technology Research System (CARS-32-21), Hainan Modern Agricul-
655
tural Industry Technology System (HNARS-08-G02).
656
Conflicts of Interest
657
The authors declare that there is no conflict of interest regarding the publication of this article.
658
Data Availability
659
The implementation code is publicly available at https://github.com/johnhamtom/FG-LCNet .
660
Upon acceptance, a representative subset of approximately 100 annotated litchi images will be
661
released to support reproducibility and preliminary benchmarking. The full dataset is being further
662
organized for future release. Before full release, the complete dataset can be obtained from the
663
corresponding authorOpen asset ↗johnhamtom/FG-LCNetpdf-raw-page:28 lines:1-81Plant phenotyping relevance matchOpenAlex · arXiv · checked 15 Sept 2026
Accurate 3D plant organ segmentation is fundamental to automated phenotyping. Existing approaches rely on annotated training data or species-specific model configurations. We present an annotation-free pipeline for 3D plant organ segmentation, combining text-prompted SAM3 segmentation with semantic neural radiance fields (NeRFs). Given only multi-view RGB images and a list of class names, our zero-shot pipeline produces semantically labeled 3D point clouds without manual annotation, per-species fine-tuning, or domain-specific preprocessing. Multi-view NeRF fusion acts as effective implicit consensus mechanism that lifts imperfect per-frame masks into accurate 3D labels. On a controlled Begonia maculata testbed the SAM3 pipeline achieves 92.6% mIoU, reaching 95.9% of the oracle upper bound established with perfect ground-truth masks. The pipeline was further evaluated on a new dataset spanning ten diverse plant point clouds reaching an average 0.856 mIoU, with leaf and pot IoU above 0.91 and 0.90 for every species, respectively. These results demonstrate that annotation-free 3D plant organ segmentation is now feasible and approaching the range of supervised methods.
Medicinal plants are critical sources of bioactive materials, yet efficient field-scale monitoring of their growth responses to water and nitrogen management remains limited. The present work evaluated Unmanned Aerial Vehicle (UAV) multispectral sensing for estimating leaf area index (LAI), underground fresh weight (UFW), aboveground fresh weight (AFW), underground dry weight (UDW), and aboveground dry weight (ADW) in Astragalus membranaceus var. mongholicus (Bunge) P.K.Hsiao ( A. membranaceus ) under 15 irrigation-nitrogen treatments spanning irrigation inputs of 15.0-30.0 mm per event and nitrogen application rates of 0–450 kg N ha⁻¹. UAV imagery and ground measurements were acquired from 45 plots at 60, 75, 90, 105, 120, and 135 days after sowing (DAS). Thirty-seven vegetation indices (VIs) were screened, and random forest (RF), support vector machine (SVM), back-propagation neural network (BP), genetic algorithm-optimized BP neural network (GA-BP), and decision-level fusion (DLF) models were evaluated with repeated random partitions. Growth responses were trait-dependent. W1N5 reached the peak final LAI (1.96), whereas W2N2 achieved the maximum UFW, AFW, UDW, and ADW (4363.94, 7406.00, 1354.50, and 2596.51 kg ha⁻¹, respectively). VI-trait associations changed with growth stage. GNDVI correlated most closely with LAI at 75 DAS (r = 0.78), and biomass-model performance was stronger during 90–120 DAS than at 60 DAS. DLF was retained in 24 of the 30 stage-specific models. In whole-season modeling, LAI was best estimated by BNDVI input with BP (R 2 = 0.96; RMSE = 0.01), whereas UDW was best estimated by DVI input with DLF (R 2 = 0.93; RMSE = 0.39). Stage-adaptive feature selection increased R 2 by up to 38.60%. Overall, the results indicate that UAV multispectral sensing, stage-adaptive VI screening, and ensemble learning provide a practical framework for monitoring A. membranaceus growth and supporting precision irrigation-nitrogen management in medicinal plant production.
Abstract Maize plant plays crucial role not only in the field of agriculture but also in global economy, since it is the third most cultivated crop across the globe. However, these plants are usually affected by various types of diseases such as blight, common rust, gray leaf spot, etc., protecting the plants from these disease is very important. This research proposes a new deep learning model for disease detection to perform well than other models. The hybrid model uses MobileNetV3 as the backbone architecture integrated with attention, fusion and head. The framework includes data preprocessing, feature extraction, classification and interpretation. We will compare the performance of this model with other models such as VGG16, ResNet50, DenseNet121, ALEXNET, etc. criteria for the final evaluation includes Accuracy, Precision, Recall, F1score, Specificity, Logloss, AUC-ROC curve. Through this proposed model we have achieved an accuracy of 98% which is high than the other models compared. The lightweight nature of MobileNetV3 enables us to implement the model in the mobile and IoT devices also. The present study contributes to the development of deep learning model in the field of agriculture, offering a efficient solution for early maize leaf disease detection.
Reproduction assets foundThe paper's maize leaf disease detection model (MAFH) was trained and evaluated entirely on a public Kaggle image dataset, which the authors explicitly declare in the Data Availability statement. No author code, trained model checkpoints, or other paper-specific assets are stated as publicly available.Dataset · publicig and real-
time datasets and in all the environmental situations.
Funding: This research received no external funding.
Disclosure statement: The authors declare no conflict of interest.
Data Availability
The datasets generated and/or analyzed during the current study are available in the CORN OR
MAIZE LEAF DATASET repository,
https://www.kaggle.com/datasets/smaranjitghose/corn-or-maize-leaf-disease-datasetOpen asset ↗Kaggle · corn-or-maize-leaf-disease-datasetpdf-raw-page:27 lines:1-34Plant phenotyping relevance matchCrossref · checked 15 Sept 2026
Abstract The timely detection of plant health status is essential to obtaining several advantages, such as increasing crop yield, reducing the toxic pollutants used in crops, enhancing healthy crops, and enhancing economic returns. A computer-aided plant status identification system enables plant health identification using plant leaves because plant leaves are the most prominent in the plant. Deep learning (DL) and convolutional neural networks (CNN) are distinguished in the area of plant health identification. Still, CNN cannot handle images of variable size and fails to extract features efficiently if the image has complex backgrounds. To avoid these problems, this research proposes a hybrid model for plant (groundnut) illness status identification using the CNN and vision transformer (HCVT). CNN is efficient in identifying local features, and ViTs are efficient in identifying global features from leaf images; leveraging these advantages enhances the model's efficacy. The HCVT model used the two datasets, groundnut and rice datasets, for experimentation. The HCVT model obtained an accuracy of 96.40% on the groundnut leaf image dataset, with 99.86% accuracy on the rice dataset. The ablation analysis performed the necessity of each component in the HCVT model. The LIME (local interpretable model-agnostic explanations) technique is utilized to comprehend the HCVT method functionality and its results showed that the HCVT model accomplished better than the present cutting-edge models in identifying plant illness and exhibiting its generalization potential. The HCVT model is affordable and widely accessible for analysing plant leaf images. Utilizing the HCVT model offers a robust and easily accessible approach for diagnosing plant diseases by analysing leaf images.
Mangrove ecosystems are frequently exposed to high concentrations of iron (Fe) in sediments, resulting in Fe accumulation in plant tissues. Although Fe is an essential micronutrient involved in several metabolic processes, its excess requires efficient mechanisms of compartmentalization and storage to maintain cellular homeostasis. Histochemical detection using the Perls reaction has usually been applied to identify ferric iron (Fe 3+ ) in biological tissues; however, the combination of this technique with ultrastructural and elemental analyses remains relatively unexplored in plant cells. In this study, we investigated Fe localization in leaf tissues of Rhizophora mangle L. (Rhizophoraceae), a dominant mangrove species, by combining complementary approaches, including Perls cytochemical reaction, transmission electron microscopy (TEM), scanning transmission electron microscopy coupled with high-angle annular dark-field imaging (STEM-HAADF), and energy-dispersive X-ray spectroscopy (EDS). Perls-positive electron-dense deposits were visualized at the ultrastructural level, and their elemental composition was further characterized by EDS analyses. Fe-containing deposits were detected in the epidermis, mesophyll parenchyma, mucilage cells, and vascular tissues, as well as in multiple cellular compartments, including plastids, mitochondria, vacuoles, cell walls, intercellular spaces, and plasmodesmata, whereas sclerenchyma cells showed no detectable Fe-containing deposits. The combination of Perls reaction with TEM, STEM-HAADF, and EDS provides a complementary approach for high-resolution visualization and elemental characterization of Fe-containing deposits at the subcellular level. This integrated methodology may facilitate the investigation of Fe distribution and compartmentalization in plant tissues under contrasting conditions of Fe availability.
Water scarcity and increasingly irregular rainfall threaten avocado production in Mediterranean regions, yet the long term physiological responses of mature trees to sustained deficit irrigation remain poorly understood. We conducted a two-year field study integrating continuous monitoring of the soil plant atmosphere continuum, drone-based multispectral imaging, canopy structural analysis, and fruit phenotyping in a mature avocado orchard subjected to three irrigation regimes. The two study years differed markedly in rainfall, providing a unique opportunity to evaluate how environmental conditions modulate tree responses to water limitation. Trees under severe deficit irrigation showed depletion of water in deeper soil layers and a flattened physiological profile, with near-zero diel variation in leaf thickness and trunk water potential, indicating minimal transpiration and decoupling of tree water status from environmental demand. Drone telemetry via NDVI detected stress during fruit growth and maturation, but not during flowering or the new summer leaf flush, revealing greater drought sensitivity at later maturation stages. Although canopy area did not differ among irrigation treatments, canopy surface roughness increased significantly under deficit irrigation, thereby identifying a novel structural indicator of drought stress. Despite large physiological differences among treatments, fruit number remained stable, while fruit weight decreased significantly under severe deficit irrigation, particularly in the wetter year, suggesting that annual rainfall modulates the trade-off between fruit retention and fruit growth. This study provides the first continuous, multi-scale characterization of avocado performance under sustained deficit irrigation in Mediterranean conditions. By integrating plant-based sensors, remote sensing, and artificial intelligence, we reveal previously undescribed stress dynamics and identify new indicators for precision irrigation management in fruit crops.
Rice leaf diseases significantly reduce agricultural yield and pose a major challenge for sustainable food production, particularly owing to the limitations associated with manual and visual inspection methods that are subjective and often ineffective in early-stage detection. This investigation proposes a new Hybrid ResConvolutional Neural Network (HyResCN-Net) for effective classification and severity identification of rice plant leaf diseases. The proposed framework integrates an Internet of Things (IoT) -based data acquisition and routing simulation using CrowWhale Energy Trust Routing (CrowWhale-ETR) for efficient data handling. Initially, preprocessing is done by an averaging filter to reduce noise. Then, plant leaves are segmented using the Eff-UNet++ method. Augmentation techniques like rotation, scaling, and color change are applied to expand the dataset. Features, like entropy with Gradient Directional Pattern (GDP), Complete Local Binary Pattern (CLBP), and histogram features, are extracted to enhance feature representation. These features are then used within the proposed HyResCN-Net model, which integrates Parallel Convolutional Neural Network (PCNN) and ResNeXt to improve discriminative learning for disease classification and severity estimation. Experimental evaluation is conducted on the Rice Leaf Bacterial and Fungal Disease Dataset. Considering a k-value of 8, the HyResCN-Net gains an accuracy of 94.258%, a True Positive Rate (TPR) of 96.479%, a True Negative Rate (TNR) of 92.898%, a precision of 91.312% and an F1-score of 93.824% compared to existing methods. The HyResCN-Net efficiently enhances rice leaf disease identification and severity analysis, supporting its applicability in precision agriculture applications.
This study assessed the integration of Unmanned Aerial Vehicles (UAVs) and satellite images (Sentinel-1 and Sentinel-2) in advanced machine-learning techniques to monitor the ABV of quinoa crops (Jacha Grano variety) across the Bolivian Altiplano. The proposed method follows a two-step procedure. First, UAV RGB images were used in photogrammetric and deep-learning (Convolutional Neural Networks-CNN) models to estimate reference quinoa ABVs at a 10 m spatial resolution from the crop canopy 3D model and classification, respectively. Secondly, several spectral and polarization/texture indices derived from Sentinel-2 and -1 images were integrated into three decision-tree-based machine-learning models (Random Forest-RF, Gradient Boosting-GB, eXtreme Gradient Boosting-XGB), and one CNN-based machine-learning model to estimate ABV. Additionally, a Stacking Model (STM) build on top of the three decision-tree-based models was considered for comparison. Model evaluation was also performed in a two-step approach. First, a 10-fold cross-validation strategy was used to highlight ABV sensitivity to Sentinel-2 and Sentinel-1 alone and in combination. Secondly, a Leave-One-Plot-Out Cross-Validation (LOPOCV) strategy was used to avoid autocorrelation between the training and evaluation dataset and therefore provided more insight into ABV mapping potential. The results showed that the combination of Sentinel-1 and Sentinel-2 features in the CNN model achieved the best predictive performance with R2 and RMSE values of 0.64 and 0.39 m3 ∙ 100 m−2, respectively. These findings highlight the potential of integrating multi-source information in advanced artificial intelligence algorithms for quinoa ABV monitoring, offering new insights toward the identification of sustainable practices across remote regions with complex socio-economic contexts.
Abstract Deep learning models have achieved near-perfect accuracy in plant disease classification within controlled laboratory settings; however, their deployment in real-world agricultural environments is severely hindered by the “deployment gap”—a critical vulnerability to environmental corruptions such as sensor noise, motion blur, and occlusion. To bridge this gap, we propose the Recurrent Active Vision Agent (RAVA), formulating the disease detection task as a Partially Observable Markov Decision Process (POMDP). Unlike passive Convolutional Neural Networks (CNNs) that process images globally, RAVA mimics the active inspection behavior of human agronomists. Our architecture integrates a lightweight ResNet-18 backbone with a Recurrent Neural Network (RNN) and a Spatial Transformer Network (STN). Driven by Proximal Policy Optimization (PPO), the agent learns a sequential policy to intelligently navigate and zoom in on informative “glimpses,” effectively bypassing background clutter. To stabilize the reinforcement learning process and enforce noise-invariant feature representations, we introduce a hybrid objective incorporating Supervised Contrastive Learning (SupCon). Comprehensive experiments on a combined PlantVillage and PlantDoc dataset demonstrate RAVA’s overwhelming superiority under extreme conditions. In a “Severe Degradation” stress test, standard ResNet-50 accuracy collapses to 26.3%, whereas our active agent maintains a robust 77.0%. Under extreme noise and occlusion, RAVA preserves 54.6% accuracy compared to the baseline’s 17.2%. Notably, this resilience is achieved with merely ∼12M parameters—significantly fewer than large-scale Vision Transformers—proving that active visual attention, coupled with contrastive learning, offers a computationally efficient and highly robust pathway for field-ready precision agriculture.
Quantitative pollen viability analysis is a critical but labor-intensive step in plant reproductive biology. Existing deep-learning Segment Anything Models (SAM) fail to reliably segment viable pollen in Alexander-stained anthers. To address this, we fine-tuned an existing Cellpose-SAM model for pollen segmentation. We integrated it into PAT (Pollen Analysis Tool), a cross-platform desktop application. PAT features instance segmentation with interactive quality control, an in-app model retraining module, and publication-ready statistical outputs. We deployed PAT in an EMS suppressor screen of semi-sterile Arabidopsis smg7-6 mutants, enabling efficient candidate prioritization for whole-genome sequencing and mapping of the candidate mutation. This screen led to the identification of a point mutation in CAP-D2 (capd2-2), a Condensin I subunit, that rescues the smg7-6 meiotic phenotype. Notably, mutation in a Condensin II subunits (CAP-D3 and CAP-H2) does not confer rescue. Further characterization suggests the capd2-2 allele is hypomorphic, showing no defects in vegetative growth, chromocenter compaction, or transposable element silencing. Collectively, we demonstrate that accessible AI tools have the potential to bridge gaps in plant phenotyping and accelerate the pace of biological discovery.
Reproduction assets foundThe paper's Data availability statement explicitly deposits the authors' PAT pollen-phenotyping tool (the software implementing the paper's computational analysis, including the fine-tuned CPSAM segmentation model support) as open source on GitHub. Note: the full repository URL in the text (https://github.com/Riha-429[Code · public17
Data availability
428
Pollen Analysis tool (PAT) is available as open source tool at Github repository (https://github.com/Riha-429
Lab/Pollen-Analysis-Tool).
430
Figure legends
431
Fig. 1. Cellpose performance on Alexander-stained anther cross-sections across varying pollen
432
densities.
433
Representative cross-sections of Alexander-stained anthers showing a range of pollen densities, from
434
low (top rows, light staining) to high (bottom rows, dense reOpen asset ↗Pollen-Analysis-Toolpdf-raw-page:17 lines:1-64Code / dataset availability confirmedOpenAlex · Crossref · checked 15 Sept 2026
Jie Zhou · Victor Hugo Moura de Souza · Lei Ju · Sebastian Eves‐van den Akker · Ji Zhou · Olaf Prosper Kranse
ArabidopsisRootMorphology / geometry measurementSegmentationRoot system architecture
Introduction Plant parasitism by sedentary plant-parasitic nematodes is a dynamic and continuously evolving process, accompanied by profound remodelling of host root system architecture across distinct infection stages. However, the physiology and anisotropic growth of Arabidopsis thaliana roots under Heterodera schachtii infection, together with complex lateral root proliferation and increasingly dense, overlapping morphology, pose substantial challenges for accurate image segmentation. Methods Here, we introduce Root-TransUNet, an optimised segmentation architecture that extends TransUNet by incorporating a composite loss function to improve boundary precision and structural continuity, as well as dual-stage strip-pooling (SP) modules to enhance elongated and directional root features. These adaptations address the unique morphological complexity of the infected root system. Additionally, we integrated Root-TransUNet into a high-throughput phenotyping pipeline and applied it to an existing dataset of ~120,000 images of 362 A. thaliana MAGIC recombinant inbred lines collected over several months of infection. By extracting root system architecture traits, including root surface area and estimated root volume across infection stages, we enabled stage-specific association analyses between host root growth and nematode performance across these genotypes. Results Root-TransUNet achieved strong segmentation performance, demonstrating improved structural continuity and boundary precision compared with widely used CNN- and Transformer-based baselines, including UNet++. Stage-specific analyses revealed that the relationship between host root traits and nematode performance changed as infection progressed. During establishment, nematode number was largely independent of initial root size and varied strongly among genotypes, whereas during the reproductive phase (10-30 dpi), greater root expansion coincided with reduced estimated nematode volume accumulation. Notably, nematode burden was largely independent of host root size before infection, indicating that root quantity was generally not a limiting factor for infection in this experiment. Discussion These results demonstrate that Root-TransUNet can robustly segment infected root systems across a wide range of nematode infection densities, providing a scalable image-analysis framework for studying plant-parasitic nematode parasitism in combination with host root phenotyping.
Reproduction assets foundThe paper analyzes a public BioImages dataset (S-BIAD2402) of ~400,000 RGB root/nematode infection images and provides authors' analysis code on GitHub; both are paper-specific, public, and actionable.Code · publicng molecular signatures, deepening our understanding of host-parasite resource allocation strategies, and establishing a foundation for the discovery of novel resistance mechanisms.
Code and data availability
Python-based source code for automating root analysis using the datasets above is accessible via our GitHub repository ( https://github.com/JieZhou1025/Root-nematode-interaction ).
Statements
Data availability statement
Publicly available datasets were analyzed in this study. This data can be found here: https://www.ebi.ac.uk/biostudies/bioimages/studies/S-BIAD2402 .
Ethics statement
The manuscript presents research on animals that do not require ethical approval for their study.
AuthorOpen asset ↗JieZhou1025/Root-nematode-interactionlines:412-424Plant phenotyping relevance matchCrossref · checked 15 Sept 2026
Overlapping cells in plant suspension-culture microscopy pose a particular challenge, for instance, segmentation because a single pixel may belong to more than one cell. Most standard instance-segmentation methods are not designed for this setting and tend to treat overlapping objects as mutually exclusive regions. We instead represent each cell as an independent full-cell instance and introduce MC-SlotNet, an architecture that separates competitive object-slot feature assignment from mask decoding. This allows multiple predicted masks to occupy the same image region. We further introduce a mask-level multiplicity-consistency loss that encourages the predicted number of masks covering a pixel to agree with the underlying cell occupancy. We evaluate MC-SlotNet on a newly annotated dataset of 53 Siraitia grosvenorii suspension-culture micrographs containing 4131 full-cell instances acquired at 4×–40× magnification. Using grouped five-fold cross-validation and an overlap-preserving evaluation protocol, we compare the method with Mask R-CNN, SOLOv2, and Mask2Former. MC-SlotNet achieves the best performance on AP50 (0.800), mAP50:95 (0.565), F150 (0.842), all-ground-truth Dice (0.761), AJI+ (0.754), overlap-region Dice (0.705), and overlap-instance recall (0.828). Its AP75 (0.649) is comparable to Mask2Former’s (0.651). MC-SlotNet also has the lowest inference time among the evaluated methods, at 1.113 s/image. These results indicate that decoding full-cell masks independently, rather than enforcing an exclusive partition of image pixels, is well-suited to instance segmentation in plant suspension-culture microscopy images with substantial cell overlap.
Abstract Plant diseases substantially reduce global crop yields, and cotton production is particularly vulnerable to field-acquired variability in symptom appearance, background clutter, and illumination changes that limit the reliability and scalability of expert visual inspection. This study aimed to develop an accurate, computationally efficient, and explainable framework for real-time cotton leaf disease recognition that is suitable for deployment on resource-constrained edge devices. Using the SAR-CLD-2024 dataset (322 RGB images captured under natural agricultural conditions across seven categories, including healthy and diseased leaves), images were preprocessed via resizing and normalization and augmented online in the training set (random rotations, flips, brightness/contrast adjustments, and random cropping). An EfficientNet-B3 backbone initialized with ImageNet-pretrained weights was fine-tuned using categorical cross-entropy loss and Adam optimization, with early stopping, checkpointing, regularization, and a fixed-seed 70/15/15 train–validation–test partition to enhance reproducibility and reduce leakage. Performance was evaluated on an independent test set using accuracy, precision, recall, F1-score, MCC, balanced accuracy, Cohen’s kappa, confusion matrix, multi-class ROC/AUC, and precision–recall analysis, alongside computational benchmarking (parameters, FLOPs, memory, and inference latency) and comparative experiments against contemporary CNN, lightweight, and transformer-based models. The model showed stable convergence over 30 epochs with a small training–validation gap, predominantly correct predictions with limited confusion among visually similar classes, consistently high precision–recall behavior under moderate class imbalance, and stable performance across repeated runs with low variability and a tight confidence interval. Grad-CAM heatmaps localized necrotic lesions, discoloration, and infected tissues while largely ignoring background, and failure cases were associated with early-stage symptoms, occlusion, shadows, and inter-class similarity. Overall, the framework provides a reproducible, interpretable, and efficient solution for cotton leaf disease classification with practical implications for trustworthy, low-latency, on-device decision support in precision agriculture.
Reproduction assets foundThe paper's Data Availability statement explicitly names the SAR-CLD-2024 cotton leaf dataset used for all experiments as publicly available on Kaggle with a direct URL. No author analysis code or trained model checkpoint is deposited.Dataset · publicntribute to the development of fully automated, scalable, and real-time smart agriculture
systems.
Declaration
Funding
Datta Meghe Institute of Higher Education and Research Wardha, Maharashtra, India
Data Availability: The SAR-CLD-2024 cotton leaf dataset used in this study is publicly
available through the Kaggle platform at: https://www.kaggle.com/datasets/pantho12/sar-cld-2024-dataset-for-cotton
This dataset includes annotated images of various cotton leaf diseases collected under diverse
environmental conditions. All data utilized in this work are freely accessible, and the data
processing methodology has been described in detail to facilitate reproducibility.
Conflict of interest The aOpen asset ↗Kaggle · SAR-CLD-2024pdf-raw-page:32 lines:1-38Plant phenotyping relevance matchEurope PMC · Crossref · checked 15 Sept 2026
Plant diseases pose a serious threat to agriculture, causing yield losses of 20 to 40 percent each year, resulting in more than 220,000 million dollars in economic damage and significantly affecting the global food supply. Traditional plant health monitoring practices involve visual inspection of plant tissue and can only detect the presence of disease once visual symptoms are already evident. This article presents AgriIDIA, an early-detection plant disease recognition system trained on datasets of 24-channel multispectral images derived from six optical filters (BlueIR, Hotmirror, K590, K665, K720, and K850) and six vegetation indices (NDVI, GNDVI, NDRE, EVI, REI, and SAVI). First, an exploratory data analysis is conducted on the diagnostic capability of the described 24-channel data representation, using 1,266 image stacks labeled with six classes (diseased/healthy papaya, diseased/healthy potato, diseased/healthy tomato). Next, using the results of the exploratory data analysis, the manuscript describes the training and cross-validation performance of AgriIDIA, with a macro-F1 score of 83.91 ± 3.42% and an accuracy of 84.00 ± 3.17% on the validation set. Finally, the performance of the trained model is evaluated on the reserved test set (N=190), demonstrating an accuracy of 81.05%, a macro-F1 score of 0.7398, and a weighted ROC-AUC of 0.9383. The results of this study suggest that the 24-channel multispectral representation has significant diagnostic potential for the early detection of plant diseases and can be used to design accessible phytosanitary methods for small-scale farmers.
Early detection of tree-seedling establishment is essential for monitoring regeneration success in coastal-dune plantations, where conventional field assessments remain labour-intensive and spatially limited. This study presents a deep-learning workflow for detecting early-stage Pinus pinaster seedlings using multispectral UAS-derived point clouds. Field surveys in the Quiaios National Forest, Portugal, mapped approximately 1500 seedlings using RTK GNSS positioning, biometric measurements, and field photographs. Multispectral imagery acquired with a DJI Mavic 3 Multispectral platform was processed through Structure-from-Motion to generate calibrated orthomosaics, terrain products, and dense point clouds. Training-data preparation combined pine-centred buffers, spectral conditioning, manual refinement and point-cloud class assignment. Point Transformer V3 models were trained in ArcGIS Pro and evaluated using field-mapped buffers withheld from model training within plantation-line areas. The Baseline high-recall model achieved 88% object-level recall at the operational threshold of at least three classified Pine-Seedling points per buffer. The refined hard-negative model retained 84% recall while reducing off-buffer detections from 243 to 41. False-negative analysis showed that omissions were associated with reduced crown diameter and limited branch development under the adopted buffer-based retrieval framework. These results support transformer-based multispectral point-cloud classification for scalable monitoring of early-stage pine regeneration in heterogeneous coastal environments.
Near-infrared imagery is essential for vegetation monitoring, precision agriculture, and environmental remote sensing, but multispectral UAV systems remain significantly more expensive and less accessible than conventional RGB imaging platforms. This study presents a lightweight artificial intelligence framework for reconstructing the NIR spectral band exclusively from the red spectral band acquired by a UAV. The proposed methodology formulates the reconstruction task as a pixel-wise nonlinear regression problem and employs a compact multilayer perceptron (MLP) containing only 609 trainable parameters, without exploiting spatial neighborhood information. The framework was developed and evaluated using 280 synchronized multispectral UAV image sets acquired with a DJI Phantom 4 Multispectral platform over a heterogeneous agricultural landscape in the Republic of Moldova. Of these, 252 image sets were used for model development, and 28 were reserved as a held-out within-mission test subset. Quantitative evaluation on a held-out test dataset from the same acquisition mission yielded a mean squared error of 0.010329, a root mean squared error of 0.101632, a mean absolute error of 0.079883, a coefficient of determination of 0.253383, and a Pearson correlation coefficient of 0.683637 between measured and reconstructed normalized NIR digital intensities. The results indicate that the model captures part of the red–NIR relationship under the evaluated acquisition conditions; however, the moderate coefficient of determination suggests that the reconstructed values are an approximation rather than a replacement for measured NIR observations. An illustrative NDVI-based assessment showed that broad spatial vegetation patterns remained identifiable. Rather than introducing a new neural network architecture, this work establishes a compact empirical baseline to investigate the practical performance and limitations of pixel-wise NIR reconstruction from a single red-band value with minimal model complexity.
This work presents a gold nanoparticle (Au NP)-boosted disposable paper-based electrochemiluminescence (ECL) biosensor for minimally invasive on-leaf in situ detection of endogenous H 2 O 2 in tomato leaves. The inherent capillary action of filter paper (FP) was employed to simplify reagent delivery, while gold nanoparticles (Au NPs) catalytically activated H 2 O 2 to generate reactive oxygen species, thereby driving the ECL signal output. This simple, low-cost paper-based platform enabled time-resolved monitoring of H 2 O 2 in stressed plants, providing a reliable in situ strategy for evaluating tomato physiology and early disease/pest warning.
Naincy Sagar · Nassim Belmokhtar · Nathalie Boizot · Ana Alves · David Chassagnaud · R Fichot · José Rodrigues · Luc Pâques
Raman / spectroscopyLeafTissuePhysiological trait estimationLeaf traitsWater status / transpiration
Phenotyping extensive populations remains a major constraint in tree breeding programmes, particularly due to the time-consuming and labour-intensive nature of conventional methods. Near infrared (NIR) spectroscopy, which is a high-throughput phenotyping method, offers an alternative solution, providing a rapid and cost-effective approach for assessing growth- and function-based traits on large numbers of trees. This study aimed to evaluate the potential of NIR spectroscopy-based models for predicting such traits in Larch. Specifically, delta carbon-13 ( δ 13 C), carbon (C), nitrogen (N), specific leaf area (SLA), leaf dry matter content (LDM), and phenolics on needles; the branch hydraulic trait (P 50 ), and lignin and hydroxyphenyl/guaiacyl (H/G) ratio on wood cores from an experimental study on Larix species were predicted using multivariate modelling, specifically, partial least squares regression. Reliable models were obtained for N content (R 2 training = 0.95, r 2 testing = 0.94), lignin (R 2 training = 0.95, r 2 testing = 0.94), and H/G ratio (R 2 training = 0.88, r 2 testing = 0.89), while moderate predictive performance was observed for C content (R 2 training = 0.79, r 2 testing = 0.79) and δ 13 C (R 2 training = 0.76, r 2 testing = 0.69). This methodological approach and its results encourage the transition from traditional laboratory methods to efficient, large-scale-based trait evaluation techniques in forestry.
Shail Bala · S. I. Harlapur · Aditya Kamalakar Kanade · M. P. Potdar · Gurupada B. Balol · V. B. Kuligod · Lingareddy Usha Rani
Field / plotLeafClassificationDisease symptoms / severity
Early and accurate detection of plant diseases is critical in precision agriculture to improve crop management and yield. Mungbean ( Vigna radiata L.) is highly susceptible to several foliar diseases, including yellow mosaic, powdery mildew, leaf crinkle, and cercospora leaf spot, which cause substantial productivity losses. Despite expanding applications of deep learning in plant disease diagnosis, systematic multi-architecture evaluation for mungbean disease classification under natural field conditions remains limited. This study addresses this gap by evaluating five state-of-the-art deep convolutional neural network (DCNN) architectures on a large-scale, field-acquired mungbean dataset that captures real-world variability across environmental conditions and disease severity levels, distinguishing it from controlled laboratory studies. A total of 5,617 original images across five classes were used. Data augmentation was applied exclusively to the training subset after stratified splitting to prevent data leakage. The dataset was partitioned into training (70%), validation (15%), and testing (15%) subsets. VGG16, VGG19, ResNet50V2, DenseNet121, and InceptionV3 were evaluated using identical transfer learning and fine-tuning protocols. Model performance was assessed using AUC-ROC, Cohen's kappa coefficient, McNemar's test for pairwise statistical comparisons, five-fold cross-validation, and Grad-CAM-based interpretability. On the independent test set, InceptionV3 achieved the highest accuracy (98.47%) and macro-F1 (98.49%), followed by VGG16 (98.36%) and VGG19 (97.89%). AUC-ROC values exceeded 0.997 for all models, confirming excellent class discrimination. Grad-CAM visualizations further confirmed that model predictions were based on biologically relevant disease symptoms. The findings demonstrate the effectiveness of deep learning for robust disease recognition under realistic field conditions and highlight the potential of AI-based diagnostic tools for crop health monitoring, precision agriculture, and decision-support systems in mungbean production.
Simonetti V, Bonato B, Guerra S, Avesani S, Semenzato L, Bulgheroni M, Gjinaj G, Ravazzolo L, Dadda M, Castiello U.
Growth chamberMultimodalStereoRootStem / branchTrackingGrowth / development / phenology
Understanding plant behaviour requires the integration of multiple phenotypic and physiological signals measured over time under controlled conditions. However, different plant signals are typically studied using separate experimental setups, limiting temporal alignment and integrative analyses.We present Mind(the)Plant, a modular experimental facility designed for the synchronized, long-term acquisition of multimodal plant data, including three-dimensional shoot kinematics, above- and below-ground volatile organic compounds (VOCs) and root imaging. Its modular architecture is designed to accommodate additional acquisition modules, such as electrophysiological signalling, as future extensions. The platform integrates a controlled growth environment with stereovision imaging, high-resolution time-of-flight mass spectrometry and custom rhizocameras. These components are connected through a unified network infrastructure that ensures synchronized acquisition and centralized data handling.We validate the performance of each acquisition module through multi-week recordings, demonstrating high-temporal stability, reliable stereovision synchronization, effective isolation of VOCs signals and robust operation of below-ground imaging. We further illustrate the analytical potential of the platform using a one-day continuous multimodal acquisition combining shoot kinematics, above-ground VOC emissions, rhizocameras observations and environmental data.Mind(the)Plant provides a novel methodological framework for studying plant behaviour, signalling and phenotypic plasticity in ecological and evolutionary research. By enabling coordinated measurements of multiple plant response modalities, the platform supports investigations of dynamic plant-environment and plant-plant interactions from a behavioural perspective.
Reproduction assets foundThe paper's data availability statement explicitly deposits data, code and processing pipelines (supporting the multimodal plant phenotyping measurements and analysis) in a public Zenodo archive with an authors' URL matching an allowed URL.Code · publicf Interest Statement
The authors have no conflicts of interest to declare.
Peer Review
The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1111/2041-210x.70411 .
Data availability Statement
Data, code and processing pipelines supporting this study are available at https://doi.org/10.5281/zenodo.22095454 ( Simonetti & Castiello, 2026 ).
References
Avesani S, Bonato B, Simonetti V, Guerra S, Ravazzolo L, Gjinaj G, Dadda M, Castiello U. Comparing proton transfer reaction (PTR) and adduct ionization mechanism (AIM) for the study of volatile organic compounds. Molecules. 2026;31(3):402. doi: 10.3390/molecules31030402.
Baluška F, LeOpen asset ↗zenodo · 10.5281/zenodo.22095454lines:482-508Plant phenotyping relevance matchOpenAlex · checked 15 Sept 2026
Automated fruit counting and yield estimation systems are necessary for efficient orchard management. This study presents a computer vision system based on video multi-object-tracking for fruit load estimation in apple orchards and provides a comprehensive analysis of the system performance under diverse scanning conditions. The system integrates fruit detection, tracking, localization within orchard, and fruit load map generation. Experiments were carried out in an experimental apple orchard containing 420 apple trees. Data was collected with two different RGB-D sensors (Azure Kinect DK and ZED 2) at three different scanning distances (125 cm, 175 cm, and 225 cm) on two different dates prior to the harvest. Comparing the two evaluated sensors, Azure Kinect provided more consistent performance across different dates. Results also show that the longer scanning distance improves accuracy due to seeing the full tree view gives better fruit counts than close partial views. Between the two dates, best results were achieved near harvest due to fruit color at this stage, achieving a Mean Absolute Percentage Error (MAPE) of 6.91 % and a determination coefficient (R 2 ) of 0.733 (using ZED2 sensor at 225 cm distance). Finally, a test comparing scanning from one or both sides of the tree row showed that bilateral scanning improved fruit load estimation at the stretch level by incorporating information from both sides of the canopy. The results of this work demonstrate the effectiveness of the video fruit tracking systems as a useful tool for automating fruit load estimation.
Standardized extraction of quantitative phenotypes from images is increasingly important across plant biology, from ecological and evolutionary studies to genetics, breeding, and functional genomics. However, as large image datasets are increasingly used for trait analysis, many biologically relevant traits, including size, shape, color, and spatial patterning, are still measured manually or using fragmented semi-automated workflows. These limitations reduce throughput, reproducibility, and accessibility, especially for researchers without computational expertise. Here, we present BioIMA, an open-source desktop tool for rapid and standardized phenotyping from biological images. BioIMA integrates foundation model-based segmentation with automated trait computation, allowing users to extract quantitative measurements from images through an intuitive graphical interface and without model training. To validate its performance, we quantified a set of knot morphological traits in two Populus species, as these measurements are typically time-consuming to perform manually. Automatic measurements showed strong agreement with manual ImageJ-based measurements (R2 > 0.95), while reducing per-image processing time by approximately 75% (from ~15 s to ~4 s). BioIMA was further applied to diverse plant datasets, including Helianthus and Rhododendron images with varying morphologies and background conditions. Although developed for plant phenotyping, BioIMA may also be extended to other biological samples where region-based size, shape, or color traits are of interest. By combining accessibility and standardization in a lightweight local application, BioIMA provides a practical community resource for image-based phenotyping in ecological and evolutionary studies.
Reproduction assets foundThe paper's own phenotyping tool BioIMA (source code, documentation, example datasets, and user manual) is publicly available on the authors' GitHub repository, directly supporting the paper's image-based trait extraction and validation analyses.Code · publicis powered by embedded models
97
currently including SAM (Kirillov et al., 2023) and mobile SAM (Zhang et al., 2023),
98
which are executed locally through ONNX Runtime for efficient inference without
99
internet connectivity. Source code, documentation, example datasets, and a user manual
100
are publicly available on GitHub (https://github.com/jingwanglab/BioIMA).101
preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
The copyright holder for this
this version posted September 3, 2026.
;
https://doi.org/10.64898/2026.08.30.747465
doi:
bioRxiv preprintOpen asset ↗jingwanglab/BioIMApdf-raw-page:4 lines:1-60Plant phenotyping relevance matchOpenAlex · checked 15 Sept 2026
Leaf hyperspectral reflectance can provide a scalable way to estimate photosynthetic capacity (Vcmax), but models trained in one species or measurement context often lose accuracy in another. This transfer problem limits the use of spectral approaches in multi-species crop phenotyping and carbon-cycle applications. Here, we tested physiology-informed inputs for leaf-level Vcmax25 retrieval using paired gas-exchange and reflectance data from wheat (C₃; n = 198) and maize (C₄; n = 81) grown under contrasting nitrogen supply. The four input configurations were raw spectra (Mod1), spectra scaled by a PPFD–absorptance proxy (Mod2), scaled spectra augmented with radiative-transfer-derived traits (Mod3), and scaled spectra augmented with a spectral coordination proxy (Mod4). Within datasets, the best models reached R² = 0.82 in wheat, 0.41 in maize, and 0.76 in the combined dataset. In a matched comparison with a common random-forest learner, the spectral coordination proxy Mod4 improved accuracy only slightly over Mod2 in wheat (RMSE −0.51%; p = 0.0058) and maize (RMSE −1.26%; p = 0.0011) but not in the combined dataset (RMSE −0.15%; p = 0.074), and the trait-based Mod3 showed no consistent benefit. When wheat models were tested on measurement dates not used in training, accuracy remained moderate (R² = 0.563; RMSE = 15.07 µmol m⁻² s⁻¹). Despite this within-dataset performance, models applied to the other species without calibration failed in both directions (negative R²), and adding source-species data did not improve prediction even when a few samples of the new species were used for calibration. These results show that physiology-informed input design provides at most small within-dataset gains, and that reliable prediction across C₃ and C₄ crops requires calibration data from the target crop.
Fei Liu · Yingjie Fan · Mingtao Zhou · Huabing Liu · Shuya Chen · Bin Wen
LeafSegmentation
Abstract Accurate leaf instance segmentation is fundamental to quantifying morphological and structural traits in image-based plant phenotyping. However, substantial variations in leaf scale, shape, and orientation, together with dense overlap and occlusion, often lead to ambiguous instance boundaries. In addition, repeated downsampling and feature reconstruction can progressively erode fine structural details, hindering contour preservation and the separation of adjacent leaves. To address these interrelated challenges, we propose DMSHA-Net, a dynamic multi-scale feature fusion and hierarchical attention network for leaf instance segmentation. Its direction-aware Multi-Scale Feature Aggregation (MSFA) encoder captures complementary horizontal and vertical contextual information across multiple receptive-field scales, thereby improving the representation of diverse leaf morphologies. The Dense Feature Aggregation (DFA) decoder selectively integrates deep semantic and shallow structural features through stage-specific attention mechanisms, where self-attention at coarse resolutions models long-range dependencies, whereas lightweight channel recalibration at high resolutions refines local structures and boundaries. The Learnable Feature Fusion (LFF) module subsequently integrates multi-level semantic and boundary features using normalized learned weights. DMSHA-Net achieves Best Dice (BD) scores of 93.17%, 85.12%, and 90.91% on KOMATSUNA, MSU-PID, and the CVPPP-A1 subset, respectively, with corresponding foreground–background Dice (FBD) scores of 98.19%, 91.02%, and 98.24%. These results indicate that DMSHA-Net achieves competitive segmentation performance across the three datasets, particularly in scenarios involving pronounced scale variation and dense leaf overlap.
In situ monitoring of plant responses to stress is one of the most challenging aspects of precision agriculture, and the dynamic control of crop growth according to fluctuating environmental factors. Although fluorescence imaging provides a nondestructive approach for monitoring stress-related biomarkers, its performance is often hindered by the low abundance of endogenous signaling molecules and strong tissue autofluorescence. Here, we report a microneedle-integrated sensing platform that enables sensitive detection of endogenous hydrogen peroxide (H 2 O 2 ) in living plants. The platform incorporates a second near-infrared fluorescent nanoprobe composed of Er 3+ -doped lanthanide nanoparticles emitting at 1550 nm and Mo-doped polymetallic oxomolybdates serving as the H 2 O 2 -responsive unit. Embedding the nanoprobe into custom-fabricated microneedles allows precise positioning on plant midribs for continuous monitoring of H 2 O 2 dynamics. Under stress conditions, the system successfully visualized spatiotemporal fluctuations of H 2 O 2 in living tomato, spinach, and tobacco plants. This work establishes a strategy for early stress diagnosis and developing universal plant health monitoring technologies.
Abstract Maize is the staple crop for millions of people in Sub-Saharan Africa, particularly for Zambia. Unfortunately, maize crops are exposed to several serious threats due to their susceptibility to foliar diseases like Maize Rust, Leaf Blight, Leaf Spot, Maize Streak Virus, and Maize Lethal Necrosis that may lead to great yield losses. Conventional methods of crop disease identification consist of field surveys that are not only subjective but also difficult to conduct for smallholder farmers. This paper presents the design and evaluation of a highly optimized version of the EfficientNet-B0 Convolutional Neural Network for the automatic detection of maize leaf diseases using maize leaf images obtained from real-world scenarios. The proposed model utilized the concept of transfer learning with ImageNet pre-trained weights and was trained on the Mendeley Maize Crop Disease (Leaf) Dataset which consists of 30,120 images in nine maize disease classes. The developed fine-tuned EfficientNet-B0 yielded 97.57% classification accuracy, macro precision of 97.61%, macro recall of 97.64%, and macro F1-score of 97.61%. From these results, it is evident that transfer learning and fine-tuning greatly boost maize disease classification accuracy while ensuring high computational efficiency. This study makes a significant contribution to precision agriculture as it offers an accurate and computationally efficient AI-based maize disease classification model, which could help smallholder farmers in early maize disease classification.
Reproduction assets foundThe paper's sole qualifying asset is the public Mendeley Maize Crop Disease (Leaf) Dataset of maize leaf images used for all phenotyping/classification measurements, explicitly declared publicly available with a URL. No author analysis code, trained model checkpoints, or other paper-specific assets are disclosed.Dataset · publiconflicts of interest to publish the paper.
Consent to Publish
All authors have read and approved the final version of the manuscript and agree to its
submission to Discover Networks.
Consent to Participate
Not applicable
Data Availability
The Mendeley Maize Crop Disease (Leaf) Dataset used in this study is publicly available at
https://data.mendeley.com/datasets/6w6gsvghfw
Clinical Trial Number
Not applicable.
Ethics Declaration:
Not applicable.
Competing interests
All authors declare no competing interests.Open asset ↗Mendeley · 6w6gsvghfwpdf-raw-page:55 lines:1-22Plant phenotyping relevance matchCrossref · checked 15 Sept 2026
Brown Rot (Monilinia spp.) and Leaf Curl (Taphrina deformans) are principal fungal diseases affecting peach (Prunus persica L. Batsch) production, lacking validated AI-based diagnostic tools in tropical highland orchards. This study presents a compact Convolutional Neural Network (3658 trainable parameters), applied identically to fruit and leaf classification, integrated with a background-removal preprocessing pipeline and evaluated through stratified 5-fold cross-validation on 800 in-situ images from four orchards in Cómbita and Choachí, Colombia. The proposed architecture achieved mean accuracies of 82.8% (fruit) and 95.3% (leaf), with AUC values of 0.87 and 0.98, and a trained model footprint of approximately 100 KB, supporting storage- and bandwidth-efficient deployment. Benchmarked against ImageNet-pretrained MobileNetV3-Small and MobileNetV2 under an identical protocol, the proposed architecture matched or exceeded MobileNetV3-Small on leaf classification despite a 257-fold smaller parameter count, and achieved comparable or lower inference latency than both larger backbones. To our knowledge, this is the first validated system for simultaneous detection of both pathogens in Prunus persica under real field conditions, combining a compact, deployment-ready architecture with an ablation-verified preprocessing pipeline. The proposed model was deployed in the DurAPP web platform, giving peach growers in tropical highland regions a practical, low-footprint diagnostic tool suited to smallholder farming conditions.
Occlusion is a major factor limiting accurate three-dimensional (3D) wheat phenotyping. In natural growth conditions, overlapping spikes, leaves, and stems often make only partial target regions visible in single-view images, hindering complete and reliable 3D reconstruction. To address this problem, this study proposes an amodal completion-assisted, sequential framework for single-view 3D reconstruction of occluded wheat. The framework first uses visible prompts to recover the complete appearance and structural cues of occluded targets, and then feeds the completed images into single-view 3D reconstruction models to generate complete 3D structures. We construct the MMWO (Multi-view Multi-instance Wheat Occlusion) dataset from MMW, which is captured under controlled indoor scenarios, by synthesizing diverse occlusion samples through organ-level cutouts, random geometric transformations, and region-constrained pasting, with annotations including visible masks, occlusion masks, and complete target images. Six representative reconstruction methods, including Direct3D, Real3D, SF3D, Spar3D, TRELLIS.2, and Hunyuan3D, are systematically evaluated. Hunyuan3D achieves the best geometric performance, with the lowest mean CD- \(L_1\) and CD- \(L_2\) values of 0.1286 and 0.0536, and the highest mean F-score of 0.5668. SF3D achieves the best rendering quality in terms of PSNR, SSIM, and LPIPS. In addition, Pix2Gestalt completion reduces the estimation errors of spike length, width, and area from 9.31%, 10.89%, and 32.23% to 4.64%, 9.70%, and 9.45%, respectively. These results demonstrate that amodal completion can effectively alleviate occlusion-induced information loss and provide more complete structural priors for single-view 3D wheat reconstruction. This study offers a feasible solution for 3D wheat phenotyping under occlusion and provides a systematic reference for applying 3D generative models to agricultural phenotyping.
Miho H, Barranco Navero D, Trujillo Navas I, Morello Parra P, Koubouris G, Trapero Ramirez C, López-Bernal Á, Oueslati A, Flores Mesas D, García López MT, Valverde Caballero P, Deger RE, Meca E, Perri E, Santilli E, Moral Moral J, Estudillo Cazorla C, Cabello Pozo D, Yousef Yousef M, Acar S, Gurbuz-Veral M, Priego Capote F, Zahiri A, Oulbi S, Márquez Pérez MI, M. Díez C.
Background Olive ( Olea europaea L.) breeding initiatives rely heavily on the extensive and correct characterisation of genetic resources to successfully achieve their goals, such as addressing climate change and emerging diseases challenges. However, the historical lack of standardised phenotyping protocols across multi-environment trials has severely hindered data interoperability and large-scale comparative analyses. Methods Within the Horizon 2020 GEN4OLIVE project, five international olive germplasm banks established a consensus-based methodological framework to systematically evaluate over 500 olive cultivars. We harmonised 14 evaluation protocols covering six fundamental dimensions: phenological and agronomic traits, abiotic stress resilience, biotic stress resilience, olive oil yield and chemical quality, table olive quality assessment, and morphological characterisation and photography. While most protocols were adapted from previously published literature to ensure ease of implementation across different facilities, novel methodologies for frost tolerance and standardised photography were developed de novo. Results The implementation of these consensus methods across five countries proved highly successful. This methodological framework enabled the generation of the largest harmonised, publicly available dataset on olive genetic resources to date, effectively making possible the correct comparation and ranking of the olive cultivars based on their specific characteristics. Conclusions This compendium of methods provides a robust, highly replicable reference point for the standardisation of olive germplasm characterisation and use of shared benchmark cultivars as an effective way for data normalization and comparation. It facilitates future global pre-breeding efforts, ensures international data interoperability, and supports the discovery of resilient cultivars to secure the future of the olive sector.
Reproduction assets foundThe article declares two paper-specific public assets: the GEN4OLIVE phenotypic dataset from evaluating over 500 olive accessions across five germplasm banks, hosted on the project's Olive Varieties Database, and a Zenodo-deposited methodological handbook (Extended Data) containing the 14 protocols, visual assessment, Dataset · publicData and software availability
The phenotypic dataset generated from the evaluation of over 500 olive varieties across the five Mediterranean
germplasm banks using this compendium of protocols and methodologies, is publicly available via the GEN4OLIVE
project repository.
• Repository: GEN4OLIVE Olive Varieties Database.
• Link: https://www.uco.es/ucolivo/gen4olive/olivevarieties (GEN4OLIVE Database, 2025).
Page 8 of 15
Open Research Europe 2026, 6:322 Last updated: 14 SEP 2026Open asset ↗GEN4OLIVE Olive Varieties Databasepdf-raw-page:8 lines:1-44Plant phenotyping relevance matchOpenAlex · checked 15 Sept 2026
Modern crop breeding demands precise organ-level analysis for trait quantification, making plant point cloud segmentation (PPCS) increasingly important. However, conventional deep learning approaches rely heavily on densely annotated datasets that are labor-intensive to acquire. Unified PPCS adaptation from distribution-shifted examples with minimal additional training remains challenging. To address this, we propose PlantC2USeg, a deep transfer learning framework featuring cross-scale consistency learning to explicitly align features across spatial scales and an information-restricted decoding strategy that prevents reconstruction shortcuts and promotes robust adaptation. The resulting pre-training enables stable few-shot generalization across species and sensing conditions, while unified fine-tuning with inherited thresholds further reduces adaptation overhead. Under full supervision on Soybean3D, PlantC2USeg achieves the highest semantic IoU and instance mWCov among compared methods, at 91.91% and 94.62%. With 20 labeled samples, it leads both metrics at 89.78% and 90.27%; with only 10 samples, it retains the highest mWCov of 83.23% while achieving 83.19% IoU. Across HR3D, 10-shot transfer to tobacco, tomato, and sorghum averages 78.41% IoU and 79.42% mWCov, while 22-shot transfer to SYAU-Maize achieves the highest IoU and mRec at 92.75% and 93.51%. Furthermore, a leading category-averaged mIoU of 85.0% on ShapeNet Part demonstrates the framework's capability to handle diverse shape variations beyond agricultural domains. These results demonstrate that PlantC2USeg reduces overall adaptation effort under distribution shifts, enabling scalable plant phenotyping and transferable 3D representation learning beyond agriculture.
To enhance crop performance, intercropping strategies leverage volatile organic compound (VOC)-driven interactions with companion plants that constitutively emit VOCs. Despite the agricultural importance, the mechanisms and kinetics of VOC-mediated sensory transduction in receiver plants eavesdropping on neighbouring non-kin emitters remain largely unknown due to a lack of appropriate non-destructive analytical tools. In this work, we employ multiplexed salicylic acid (SA) and H 2 O 2 nanosensors in Brassica rapa subsp. Chinensis (pak choy) plants to visualize, in real time, reactive oxygen species (ROS) and SA signal transduction following exposure to constitutively-released VOCs from neighbouring aromatic plants-namely sweet basil and spearmint. Unique emitter-specific temporal signatures of ROS and SA were observed in receiver pak choy: sweet basil VOCs induced concomitant generation of ROS and SA at 30 min, whereas spearmint VOCs triggered SA production at 30 min, followed by ROS accumulation. The temporal data enabled the formulation of a diffusion model that quantifies the VOC perception threshold that triggers the distinct early ROS and SA signalling. Transcriptomics analysis at 2 h revealed that both emitters evoke largely distinct changes in pak choy, likely stemming from variations in speed and sequence of the early signal transduction, leading to different phenotypic outcomes. Intercropping with sweet basil led to enhanced pak choy biomass, stress resilience and secondary metabolite accumulation, whereas spearmint as companions had a limited impact. Our study captures in real time, the VOC-induced rapid signalling in receiver plants and its ensuing effect on growth. These nanosensor-enabled findings represent an important advance in deciphering how emitter-specific volatile cues are integrated into plant responses, guiding rational selection of beneficial companion plants for improved yield and nutritional profiles in sustainable agriculture.
Urban air pollution may alter plant metabolism long before visible damage becomes apparent. Raman spectroscopy was evaluated as a rapid, non-destructive approach to resolve these biochemical adjustments. Mature Quercus ilex L. trees were sampled along a well-defined pollution gradient in Tuscany (Italy), spanning high, intermediate, and low levels of NO₂ and PM₁₀. Leaf Raman spectra revealed coordinated modulation of primary and secondary metabolism. Pigment-related bands (chlorophylls and carotenoids) increased toward the most polluted site, while inducible flavonoid signals showed site-dependent variation consistent with oxidative pressure in superficial tissues. These patterns were consistent with destructive biochemical analyses and chlorophyll fluorescence measurements, which indicated acclimation rather than photoinhibition damage. A composite Raman index showed a close site-level association with NO₂ exposure, suggesting that nitrogen-related urban pollution was the main exposure component linked to the observed metabolic response. Overall, Raman spectroscopy captures the chronic metabolic imprint of urban air pollution in Q. ilex, resolving coordinated pigment reinforcement and defensive activation without sample destruction. This approach provides a rapid and scalable framework for linking atmospheric chemistry to plant functional status in biomonitoring applications.
Agriculture is being revolutionized through the integration of cutting-edge technologies that support efficiency, productivity, and sustainability. Recent trends in transforming traditional agriculture into smart agricultural systems through the application of Artificial Intelligence (AI), IoT, and drone technology are gaining popularity worldwide. In this context, the current study developed an integrated artificial intelligence-based farming system incorporating Internet of Things sensors, drones, deep learning, and web/mobile applications for timely monitoring of crop and aquatic health. Data were collected using IoT sensors and drones equipped with high-resolution cameras to monitor soil and water parameters, including temperature, humidity, and pH, under various climatic conditions. A dataset of 70,000 images was used for system training, validation, and testing with an 80:10:10 split, together with three months of IoT sensor data. Models including YOLOv8, CNN, Faster R-CNN, ResNet50, MobileNet, EfficientNet, DenseNet, LSTM, and Random Forest were used for pest and disease detection, fish classification, fish disease detection, shrimp disease detection, and monitoring of climatic factors. Data pre-processing included denoising, normalization, missing-value handling, and feature extraction. The designed model showed reliable performance across the tasks, with 88.4% accuracy for shrimp detection, 92.1% for pest detection using YOLOv8, and 93.4% accuracy for plant disease detection using the CNN model. Overall performance was recorded at 97% accuracy, with high precision, F1-score, and mAP, and an RMSE of 1.2 for sensor-based prediction and validation. The current findings indicate the potential of using AI, IoT, and drone technologies to detect biotic and abiotic stresses during farming and support a sustainable agricultural system.
Abstract Agricultural pest and disease monitoring plays a vital role in ensuring crop productivity, reducing pesticide consumption, and promoting sustainable agricultural development. Although deep learning techniques have achieved remarkable success in plant health diagnosis, many existing models remain computationally intensive and are difficult to deploy on resource-constrained edge devices used in practical agricultural environments. To address these challenges, this study proposes a practical lightweight deep learning framework based on an improved ShuffleNetV2 architecture for real-time maize pest and disease recognition under complex field conditions.The proposed model incorporates the Ghost module to reduce redundant feature generation, the Efficient Channel Attention (ECA) mechanism to enhance feature representation, and the HardSwish activation function to improve nonlinear learning capability while maintaining computational efficiency. Extensive experiments were conducted on a maize pest and disease dataset containing multiple disease and pest categories collected under natural field conditions. Experimental results demonstrate that the proposed model achieves superior recognition accuracy while significantly reducing model parameters and computational complexity compared with several mainstream lightweight convolutional neural networks.The results show that the proposed method achieves an accuracy of 93.00%, a recall of 92.76%, and an F1-score of 92.42%, while maintaining extremely low computational cost (0.03 GFLOPs) and model size (1.16 MB). Furthermore, the proposed model was successfully deployed on a Raspberry Pi platform, demonstrating excellent real-time inference capability and low computational resource consumption. The framework is suitable for practical agricultural applications, including intelligent crop monitoring, UAV-assisted field inspection, and mobile diagnostic systems. By enabling rapid and accurate in-field identification of maize pests and diseases, the proposed approach supports timely crop protection decisions, reduces unnecessary pesticide application, and contributes to sustainable agriculture through practical edge-AI deployment.
Hyperspectral sensors have emerged as a promising approach in the study of plant diseases. The objective was to distinguish between healthy and inoculated seeds, and also to distinguish between genera of plant-pathogenic fungi in soybean seeds, using hyperspectral sensors combined with machine learning. The experimental design was a fully randomized factorial design with six algorithms (Simple Logistic Regression, Support Vector Machine, Artificial Neural Network, Random Forest, REPTree and J48 decision trees) and four phytopathogens (Sclerotinia sclerotiorum, Macrophomina phaseolina, Rhizoctonia solani, and Colletotrichum sp.) plus the control. Spectral analysis of the seeds was performed using a spectroradiometer (Ocean Optics) consisting of two sensors: NIR and Flame, covering the spectrum from 350 to 2500 nm. It was possible to distinguish between healthy and inoculated seeds, as well as identify the type of phytopathogen, based on each spectral signature. The Simple Logistic Regression and Support Vector Machine algorithms performed best. Hyperspectral sensors combined with machine learning constitute a promising tool for the detection of phytopathogens in seeds, enabling rapid and non-destructive analysis. This promising tool could serve as a complementary alternative to traditional diagnostic methods, which, although accurate, are time-consuming and rely on specialized labor.
Key message Yield-Graph enables accurate maize yield prediction from incomplete multi-stage phenotypic and environmental data by modeling higher-order environment-trait interactions, with robust applicability across growth stages, regions, and crop species. Accurate yield prediction before maize harvest is crucial for advancing agricultural management and ensuring food security. Unlike conventional approaches that rely on traits from a single growth stage, this study models multiple traits across different developmental stages, all targeting final yield, thereby uncovering their stage-specific contributions and demonstrating the feasibility of early yield prediction. We introduce Yield-Graph, an innovative framework that evaluates phenotypic data at distinct developmental stages for yield prediction. The method employs a bipartite graph structure to impute missing trait values at each stage and leverages a hypergraph attention mechanism to capture high-order sample relationships. Comprehensive benchmark experiments demonstrate that Yield-Graph matches the top-tier predictive accuracy of exhaustively optimized tree models. Moreover, the framework exhibits strong robustness across growth stages, high adaptability to regional variations, and effective generalization across datasets. These findings highlight the potential of graph-enhanced multi-stage modeling for early-stage yield prediction, offering a scalable solution for precision agriculture and intelligent crop management.
Laiyi Fu · Hongbo Liu · Yanbo Han · Shunkang Ling · Hongming Zhang · Fei-Yue Wang · Danyang Wu · Hequan Sun
MultimodalSegmentation
Crop phenotyping is crucial for advancing plant breeding, yet remains a significant challenge. Manual approaches are labor-intensive and do not scale to the analysis of large datasets, while computational methods like Vision-Language Models (VLMs) lack the adaptability for fine-scale spatial reasoning and diverse phenotyping scenarios. To bridge the gaps, we present iPheno, a fully open-source, domain-specialized multimodal VLM for fine-scale crop phenotyping. A key innovation of iPheno is its dual-aware architecture. First, a spatial-aware feature extractor samples mask regions into local K-Nearest Neighbors (KNN) graphs and enables fine-scale analysis of arbitrary-shaped regions; second, a task-aware Mixture-of-Experts (MoE) routing mechanism activates specialized modules for each phenotyping task. To train iPheno and achieve rigorous benchmarking, we constructed iPheno-120K, a large-scale high-precision dataset designed for multiple phenotyping tasks. Evaluations on iPheno-120k test set and other publicly available datasets showed that iPheno outperformed all fine-tuned baselines, by improving F1-score by 17.1% (LLaVA-1.6-13B) to 28.9% (MiniCPM-o-9B), while achieving the highest inference speed and memory efficiency. A web server (https://ipheno.ai4bread.com), a mobile application (www.ipheno.cn), and a stand-alone PC client (https://github.com/2997029323/iPheno-PC-Client) are available for iPheno.
High-throughput phenotyping of lettuce seedlings is highly prone to background confusion because the seedlings are small, have weak textural features, and exhibit spectral reflectance similar to that of the substrate. Traditional single-visual-modality approaches struggle to achieve reliable structural and physiological characterization simultaneously under the repetitive backgrounds and dense arrangements typical of greenhouse tray cultivation. To address these challenges, we establish a multimodal 3D phenotyping framework tailored for controlled agriculture environments, enabling the quantification of structural and physiological characteristics of lettuce seedlings. This framework is based on an unmanned ground vehicle (UGV) platform integrating a RGBD camera and a quad-band multispectral sensor which are rigidly coupled and synchronously triggered. An alignment module based on established feature matching algorithm is introduced to register the misalignment between source multispectral and RGBD images. Subsequently, we design a novel dual-backbone instance segmentation network, MS-SegNet, to enhance segmentation accuracy by hierarchically fusing geometric information with multispectral features. A robust 3D metric pose estimation pipeline, incorporating standard SfM initialization, scale recovery, and generalized ICP refinement, is constructed to generate 3D point clouds with spectral attributes and semantic labels. Finally, key structural and physiological phenotype parameters of each seedling are calculated based on the 3D semantic multispectral point clouds. Experiments demonstrate that MS-SegNet achieves significant advantages in instance segmentation of lettuce seedlings with mAP@50:95 = 0.854. The metric 3D pose estimation pipeline exhibits reliable performance under complex controlled conditions. The quality of the 3D reconstructions is indirectly validated through downstream structural trait extraction. The estimated seedling height and crown width show high correlation with manual measurements, achieving R 2 values of 0.8379 and 0.918, and RMSE values of 10.94 mm and 11.56 mm, respectively. Overall, by systematically integrating these adapted components with the novel segmentation architecture, this framework achieves stable performance improvements in 3D reconstruction, instance segmentation, and phenotypic analysis under greenhouse conditions. It provides a scalable, integrated technical solution for non-destructive, high-throughput phenotyping of crop seedlings in controlled environments.
Accurate plant disease detection remains challenging when using single-modality data, which fails to capture comprehensive disease-related features. However, many existing studies rely on pixel-level classification or prior plant segmentation and lack explicit modeling of cross-modal interactions, limiting their ability to distinguish between healthy and diseased plants. This study presents a spatial–spectral fusion deep learning framework (S2-PDD) that fuses uncrewed aerial vehicles (UAV)-based RGB imagery and hyperspectral-derived feature representations for plant-level detection of potato diseases (blackleg and Potato virus Y (PVY)), employing early fusion (modalities combined at the input stage) and middle fusion (features integrated at intermediate stages within the model backbone) strategies. The multimodal fusion models were compared against single-modal models and an existing S2ADet model. Model performance, assessed through five-fold cross-validation, demonstrated that multimodal models integrating RGB and vegetation index features achieved the highest mAPs of 86.65 ± 1.71 (%; E-RV model) and 85.74 ± 1.96 (M-RV), respectively. These mAPs were higher than those of all single-modal models, including the RGB-only (83.21 ± 1.46) and hyperspectral-only models (PCA features: 79.71 ± 1.45; vegetation index features: 85.31 ± 2.36). They also exceeded mAPs of multimodal models combining RGB with PCA features (early fusion: 83.00 ± 2.81; middle fusion: 83.11 ± 2.46; S2ADet: 84.04 ± 2.73), regardless of the fusion strategy. The superior performance highlights that vegetation index features provide strong class separability compared to other hyperspectral representations. The proposed models achieved strong plant-level detection performance, with AP of 78.65 ± 4.14 (E-RV) and 77.40 ± 3.46 (M-RV) for blackleg disease, as well as 84.82 ± 3.59 (E-RV) and 83.28 ± 3.86 (M-RV) for PVY. These results demonstrate the potential of UAV-based multimodal sensing for disease monitoring in cropping systems. A potato plant disease detection dataset was constructed and made publicly available, containing paired RGB and hyperspectral image tiles with bounding box annotations. The code is available at https://github.com/Tim-Agro/S2-PDD.
Rice panicle blast detection is an important task in plant disease phenotyping. Field-based detection remains challenging because infected spike regions are often small, sparse, elongated, and affected by overlapping panicles, complex backgrounds, and variable illumination. In this study, we propose RPB-YOLO11, a lightweight YOLO11-based detector designed for rice panicle blast detection. The model uses a Lightweight Ghost Backbone (LGB) to reduce redundant computation. It uses Anisotropic Axial Stripe Attention (A2SA) to represent elongated panicle structures. It also uses Focal Multi-Scale Attention (FMSA) for multi-scale feature refinement and Adaptive Geometric Shape IoU (AGS-IoU) for geometry-aware localization. The model was trained and evaluated on a rice panicle image dataset containing 1,055 training images, 69 validation images, and 169 test images. On the test set, RPB-YOLO11 achieved 76.09% mAP50, 45.44% mAP50-95, 73.98% precision, and 72.75% recall with 6.21 GFLOPs. Compared with the YOLO11n baseline, it improved mAP50, mAP50-95, precision, and recall by 2.73, 2.12, 1.64, and 2.11 percentage points, respectively. An Android-oriented inference application supports local image inference, detection visualization, class counting, and diseased-panicle incidence estimation. These results suggest that RPB-YOLO11 provides a practical approach for image-based rice panicle blast survey.
Reproduction assets foundThe paper links a public Hugging Face dataset used to establish the rice panicle blast detection dataset and a public GitHub release (data availability statement) containing the study's datasets/models.Dataset · publicsites, cultivars, growth stages, imaging conditions, and disease severities are still needed to evaluate generalization more fully.
Statements
Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://github.com/XuzheYang2Doc/RPB-YOLO11/releases/tag/rpb-yolo11 .
Author contributions
XY: Data curation, Formal analysis, Investigation, Methodology, Software, Validation, Visualization, Writing – original draft, Writing – review & editing. XZ: Conceptualization, Methodology, Visualization, Writing – original draft, Writing – review & editing. CX: Formal analysisOpen asset ↗XuzheYang2Doc/RPB-YOLO11 · rpb-yolo11lines:639-658Plant phenotyping relevance matchCrossref · checked 14 Sept 2026
Published1 Sept 2026Computers and Electronics in Agriculture
Accurate and efficient monitoring of tea plant growth parameters via remote sensing is essential for precision plantation management. However, spectral indices relying solely on reflectance often exhibit limited sensitivity in capturing complex tea canopy characteristics. This study developed a data-driven framework integrating spectral reflectance, frequency-domain harmonic components, and spatial texture features to construct tri-feature fusion indices (TFIs) and establish machine learning and deep learning models for tea growth monitoring. Ten-band multispectral imagery was acquired using a UAV alongside synchronous field measurements of leaf and plant biomass and nitrogen accumulation. TFIs were constructed through exhaustive feature combinations and optimized via a data-driven search strategy. Subsequently, random forest (RF), multilayer perceptron (MLP), convolutional neural network (CNN), and transformer models were evaluated using a leave-one-site-out cross-validation (LOSO-CV) strategy. The selected TFIs showed strong associations with tea growth parameters within the investigated dataset, with R2 values up to 0.63 and 0.62 for leaf dry matter and leaf nitrogen accumulation, respectively. Models incorporating selected TFIs achieved cross-validated R2 values of 0.56 for leaf dry matter (MLP), 0.59 for plant dry matter (MLP), 0.73 for leaf nitrogen accumulation (MLP), and 0.68 for plant nitrogen accumulation (CNN). These models exhibited competitive predictive performance comparable to RF, although no statistically significant differences in mean absolute error were observed under site-held-out evaluation. Furthermore, model-derived spatial maps provided insights into fine-scale spatial heterogeneity and potential interannual variations in tea growth parameters across representative plantations from 2024 to 2025. Overall, this study provides a UAV-based framework for tea growth parameter estimation by integrating multi-domain information without requiring additional environmental observations.
Fruit quality is a critical determinant of economic returns in pear production, and maintaining an appropriate fruit load (FL) is essential for achieving high yield and quality. As a direct indicator of canopy photosynthetic capacity and assimilate supply, leaf number constitutes the key biological basis of reasonable FL determination under the leaf-to-fruit ratio concept. However, accurate and efficient estimation of leaf number in mature pear trees remains technically challenging, limiting its practical use in precision FL regulation. Here, we propose a data-driven framework for leaf number and reasonable FL estimation by integrating 3D point cloud-derived canopy structure with machine learning. A pipeline for extracting 3D architectural traits was developed and implemented in the software tool FTPCT, enabling rapid and standardized trait acquisition. Through correlation analysis, multicollinearity diagnosis, and variance inflation factor screening, five key traits strongly associated with leaf number were identified and incorporated into five machine learning models optimized using Bayesian optimization. Among them, the optimized random forest regression model achieved the highest and most stable performance, with R 2 of 0.85, RMSE of 239.74, and MAE of 149.26 for test dataset. SHAP analysis identified tree crown volume as the dominant contributor to leaf number estimation. Field validation demonstrated that FL regulation guided by the proposed framework significantly improved fruit weight and size without reducing yield compared with conventional practices. Notably, the proposed approach avoids explicit leaf-level reconstruction and relies on less canopy-scale traits, substantially reducing data requirements and computational cost, and thereby offering strong potential for rapid, field-deployable FL regulation in large-scale orchards.
Reproduction assets foundThe paper's phenotyping analysis assets are the authors' publicly released LeafNumPred source code and trained models, and the FTPCT software for 3D trait extraction from pear tree point clouds. Phenotype/point-cloud datasets are only available on request.Code · public. Supplementary data
The following is the Supplementary data to this article:
Multimedia component 1
mmc1.docx (1.6MB, docx)
Data availability
Data will be made available on request. Anyone who wants to obtain other public data can contact us at taost@njau.edu.cn. The source codes and models have been made publicly available at https://github.com/Zhang-Fanhang/LeafNumPred, and the FTPCT software has been released at https://github.com/Zhang-Fanhang/FTPCT/tree/Installation-package.
References
1.Tao S., Khanizadeh S., Zhang H., Zhang S. Anatomy, ultrastructure and lignin distribution of stone cells in two Pyrus species. Plant Sci. 2009;176:413–419. [Google Scholar]
2.Zhang F., Wang Q., Yuan K.Open asset ↗Zhang-Fanhang/LeafNumPredhtml-lines:284-315Code · publical variations [34,35]. The method for calculating these traits are shown in the Supplementary information 1.
2.5. Software implementation for 3D trait extraction (FTPCT)
To facilitate efficient and standardized extraction of canopy structural traits from point cloud data, we used a standalone software tool, FTPCT (available at: https://github.com/Zhang-Fanhang/FTPCT/tree/Installation-package), which integrates the trait extraction procedures applied in this study. The software provides a graphical user interface, enabling users to process tree-level point cloud data and extract key 3D structural traits without requiring advanced programming skills.
FTPCT implements a series of predefined proOpen asset ↗Zhang-Fanhang/FTPCThtml-lines:138-149Plant phenotyping relevance matchOpenAlex · checked 14 Sept 2026
UAV-based multi-view reconstruction is an important approach for high-precision, non-destructive 3D crop phenotyping. However, in greenhouse environments, UAV image acquisition is often restricted to sparse viewpoints because of UAV-induced airflow disturbances and the structural complexity of the greenhouse, which severely hinders accurate 3D phenotyping. To address this challenge, this study develops a task-driven phenotyping framework for constrained UAV viewpoints in greenhouse environments, integrating a vision-triggered flight planning strategy with an improved sparse-view 3DGS pipeline, termed SparseBerry-3DGS, for multi-view image acquisition and 3D phenotyping of greenhouse strawberries in GNSS-denied environments. Specifically, 3D Gaussian Splatting (3DGS) is improved by incorporating flow-guided initialization, depth supervision, and an adaptive pruning strategy, which effectively alleviate geometric collapse and floating artifacts under sparse-view conditions. Furthermore, sequential semantic masks generated by SAM2 are utilized to guide the segmentation of strawberry point clouds, thereby reducing background interference and segmentation errors. Experimental results show that the vision-triggered flight strategy enables stable capture of 16 surrounding images for each target fruit. Under sparse-view conditions, SparseBerry-3DGS improves reconstruction stability, with the average peak signal-to-noise ratio (PSNR) reaching 18.25 dB, corresponding to an 18% improvement. The SAM2-based segmentation module achieves high accuracy, with the mean intersection over union (mIoU) above 0.95. Geometric evaluation based on strawberry longitudinal diameter yielded an of 0.88, supporting the accuracy of fruit-scale geometric reconstruction. For weight estimation, five-fold cross-validation yielded an of 0.90 and an RMSE of 3.62 g, showing better predictive performance than models based on 2D projected area and standard 3DGS point clouds. This study provides a new approach for high-throughput, non-invasive digital crop phenotyping in greenhouse horticulture.
Plant counting functions as a central role in the quantification system of plant phenotyping. Over the years, the field has evolved through successive technological paradigms, and its modern form has been strongly influenced by advances in visual counting from computer vision. In this review, we synthesize the historical origins of plant counting, assess its current fragmented landscape, and outline a roadmap toward standardized, universal plant counting systems. We propose a coherent conceptual framework— the four-level hierarchy of plant counting , which characterizes the environment, platform, sensor, and counting entity in biological organization —grounded in the need of high-throughput plant phenotyping. This framework aims to guide the development of plant counting systems that are not only accurate on individual dataset, but also reusable, comparable, and trustworthy across modern plant phenotyping scenarios. We argue that existing plant counting approaches are constrained by species-specific designs, sensor-dependent assumptions, and local, region-bound assessments. We hope this review can serve as a reference for building cross-species, cross-modal, and cross-scale visual plant counting systems.
Problem: Agriculture plays a pivotal role in the Indian economy, where crop production quality and quantity directly impact the livelihoods of millions. Climate variability, emerging plant diseases, and improper pesticide application contribute significantly to agricultural losses. Early and accurate disease detection is crucial for mitigating crop damage and ensuring food security. Methodology: This study presents a novel deep learning framework based on EfficientNet architecture, enhanced with Progressive Fine-Tuning Strategy for automated plant disease detection. The proposed methodology was evaluated on two benchmark datasets: the Plant Village dataset comprising 20,639 images of tomato, potato, and bell pepper with 15 disease varieties; and a drone-captured rice plant dataset containing 4432 samples from public repositories. The model’s performance was assessed using multiple metrics, including classification accuracy, precision, recall, and F1-score. To strengthen the validation of high accuracy results, additional statistical analyses like class imbalance ratio, Entropy, Chi-Square test, convergence curve, ANOVA test, Tukey’s post hoc HSD test, confidence interval, Cohen’s Kappa result, fold-wise dispersion analysis, mean, std deviation are included in the manuscript. Result: Experimental results demonstrate that the fine-tuned EfficientNetV2-B1 architecture achieved exceptional performance with 99.7% classification accuracy on the PlantVillage dataset and 99.03% accuracy on the drone-based rice disease dataset, significantly outperforming existing state-of-the-art transfer learning models. Model explainability techniques further validated the reliability and interpretability of the predictions, confirming the model’s focus on disease-relevant features.
Automated plant disease classification from leaf images demands models that jointly achieve high accuracy and efficient training convergence. Standard deep learning approaches process images through a single feature pathway, limiting their ability to capture diverse visual manifestations such as color changes, texture patterns, and spatial distributions. This paper introduces DMCNNA-FBVL, a framework integrating two complementary innovations: (1) a Deep Multi-Component Neural Network Architecture (DMCNNA) employing three specialized pathways processing color-space statistics, texture descriptors, and raw image features before fusing them through SoftMax-weighted aggregation; and (2) Feedback-Based Validation Learning (FBVL), a training strategy that periodically blends validation-set gradients into weight updates to accelerate convergence. Experiments on the New Plant Diseases Dataset (87,000 images, 38 classes) show that DMCNNA-FBVL achieves 98.7% accuracy, 98.8% precision, 98.6% recall, and 98.7% F1-score, outperforming ResNet-50 by 3.2 percentage points ( 𝑝 < 0 . 0 0 1 ). The primary reported metrics are computed exclusively on an independent 10% hold-out test set, whereas the separate 10% validation partition is used during training for FBVL gradient blending and does not contribute to final test evaluation. Five-fold cross-validation confirms stability (98.7% ± 0.15%). Ablation experiments confirm additive gains, while FBVL reduces wall-clock training time by 15% through faster convergence.
Reproduction assets foundThe paper's plant disease classification experiments use the New Plant Diseases Dataset, which the authors state is publicly available on Kaggle. The authors' code and trained models are only promised 'upon acceptance' (no public URL), so they do not qualify as public assets.Dataset · publicral
monitoring systems.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
Data availability
The New Plant Diseases Dataset used in this study is publicly available on Kaggle (https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset). Code and trained models will be made available upon
acceptance.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit
sectors.
Acknowledgments
[Removed for double-blind review.]
CRediT authorship contribution statement
[Removed fOpen asset ↗Kaggle · new-plant-diseases-datasetpdf-raw-page:19 lines:1-59Plant phenotyping relevance matchEurope PMC · checked 14 Sept 2026
Published1 Sept 2026Journal of the Royal Society, Interface
LiDAR / point cloudFlowerArchitecture / morphology / geometry
The mechanical dynamics of poricidal stamens play a crucial role in buzz pollination, but they have only been studied in a few species showing comparatively less curvature. This paper presents an integrated experimental-computational approach to characterize the dynamic behaviour of the morphologically more complex stamens of Medinilla magnifica. An experimental set-up employing a custom-built shaker and a laser triangulation sensor is developed to measure the stamen motion with high spatial and temporal resolution. The resulting displacement transmissibility reveals three distinct resonances, with the second one falling within the reported frequency range of pollination buzzes. Finite-element and multi-body models are developed and validated against experimental results. The models further reveal how the stamen morphology gives rise to its complex dynamic behaviour. The developed method enables precise parametric modelling of stamens and provides new insight into the mechanical basis of pollen release, thereby offering a framework for studying buzz pollination dynamics across plant taxa.
Abstract In Bangladesh, the potato (Solanum tuberosum L.) stands as an indispensable food and cash crop, deeply intertwined with national food security, rural livelihoods, and the broader agricultural economy. However, foliar diseases such as early blight and late blight frequently precipitate substantial yield losses and quality degradation when not identified and mitigated during the nascent stages of infection. Contemporary diagnostic paradigms remain predominantly manual and visual, relying heavily on agricultural professionals, which is often inefficient and inaccessible for remote farmers. While deep learning has demonstrated remarkable efficacy in automated plant disease recognition, existing methodologies frequently lack interpretability, disease severity quantification, and real-world field applicability. This paper introduces a comprehensive, interpretable deep learning-based framework utilizing EfficientNetV2-B0 for classifying potato leaf images into healthy, early blight, and late blight categories. By integrating Gradient-Weighted Class Activation Mapping (Grad-CAM), the model achieves high transparency, highlighting critical prediction regions. Furthermore, a severity assessment module estimates infection percentages, providing actionable treatment recommendations, ultimately enhancing agricultural decision-making.
The automation and standardization of hyperspectral imaging, particularly at high spatial resolution, are essential for advancing plant phenomics and supporting diverse plant science research. We developed HyperBird, a hyperspectral microscopic imaging robot, to automate the acquisition of up to 351 leaf-disc samples in a single tray within approximately 2.4 h and thereby support scalable plant experiments. System calibration established a spatial resolution of 24.3 μ m (full width at half maximum; FWHM), a spectral resolution of 1.78 nm (FWHM), and a depth of field of 4.53 mm, producing hyperspectral data cubes of 2195 × 2000 × 950 pixels across the 400–1000 nm spectral range. System-level and biological-sample consistency assessments confirmed stable spectral measurements during extended scanning sessions and across independently prepared trays. Thermal evaluation confirmed that the 150 W illumination design introduced minimal sample heating during scanning. HyperBird was first evaluated using grape downy and powdery mildews, where no consistent measurable effect of repeated imaging on pathosystem development was detected under the tested experimental conditions. Subsequently, HyperBird was used to characterize spatiotemporal spectral progression in grapevine leaves inoculated with Plasmopara viticola . An automated regional spectral tracing pipeline was developed to isolate spectra from retrospectively traced regions and compare them with whole-leaf averaged spectra across 0–9 days post-inoculation (DPI). Categorical mixed-effects modeling showed that these spatially resolved regional spectra exhibited significant disease-associated spectral change by 5 DPI, with a sharp transition at 6 DPI, whereas whole-leaf spectra showed delayed and weaker disease-aligned changes beginning at 7 DPI. These results demonstrate that HyperBird enables high-throughput, spatially resolved hyperspectral imaging for quantifying localized plant disease progression and provides a scalable platform for studying spectral biology across plant phenotyping applications.
Reproduction assets foundThe paper's Data Availability statement points to a public GitHub repository containing the authors' code and processed data supporting the phenotyping analyses; raw hyperspectral image data are request-only.Code · publicData Availability
857
The code and processed data supporting the findings of this study are available in the GitHub
858
repository at https://github.com/jy773Cornell/HyperBird-Robot. Raw hyperspectral image
859
data are available from the corresponding author upon reasonable request due to file size and storage
860
constraints.
861
Supplementary Materials
862
Supplementary materials accompany this article as a separate document (supplementary.pdf).
863
Supplementary Figure S1. Representative GSAM-based segOpen asset ↗HyperBird-Robotpdf-raw-page:39 lines:1-75Plant phenotyping relevance matchCrossref · checked 14 Sept 2026
Published1 Sept 2026Computers and Electronics in Agriculture
Aboveground biomass density (AGBD) is a key indicator in agricultural systems, directly reflecting crop carbon storage potential and yield levels. The World Food Studies (WOFOST) model is widely used for crop growth simulation due to its process-based interpretability. However, its application is limited by complex calibration demands and struggles with spatial heterogeneity. To address these limitations, this paper proposes an assimilation system that integrates Synthetic Aperture Radar (SAR) data into a modified WOFOST model, which requires simple calibration. WOFOST is run in potential production mode, which assumes ideal conditions to reduce input data requirements. Before assimilation, phenology-related temperature sums are aligned with phenology and meteorological data to match local growth stages. Two quantities are then estimated and updated in the model through data assimilation by minimizing the difference between SAR-derived and simulated AGBD. These quantities are the proposed yield reduction factor, which represents the proportional decrease in potential CO 2 assimilation under stresses, and the initial total dry weight at sowing. Validation experiments were conducted using multi-year cotton datasets from two farms in Georgia, USA, differing in whether irrigation was applied. Compared to WOFOST simulations and evaluated against in situ AGBD measurements, the assimilation results improve agreement and reduce error (approximately 43% RMSE reduction at the rainfed site and 15% at the irrigated site). It also delivers spatial maps of biomass, together with model-derived yield and harvest-index diagnostics, and remains operational under frequent cloud cover where optical observations are sparse. This SAR-based assimilation strategy reduces calibration demands, providing a novel and practical pathway to extend WOFOST applications to diverse agricultural scenarios.
Wu F, Liu S, Zohner CM, Townsend PA, Crowther TW, Peñuelas J, Yang D, Yang N, Dong T, Xu W, Wang Z, Liu X, Dai G, Dong J, Durán SM, Schneider FD, Zeng Y, Cornelissen JHC, Kattge J, Wu J, Asner GP, Cavender-Bares J, Reich PB, Yan Z.
Global trait axes reveal overarching dimensions of plant functional variation. However, how these dimensions are spatially organized within and across forest types remains unclear. We combined drone-based full-range imaging spectroscopy with crown-level measurements of 16 physiological, morphological and biochemical traits across temperate, subtropical and tropical forests in China to enable spatially-explicit trait mapping. Through site-training scenario, leaf-to-canopy scaling and spectral-domain modelling tests, we find that reliable canopy trait retrieval depends not only on trait and spectral coverage, but also on preserving trait-spectral relationships across sites and scales. Spectral predictions recovered observed multivariate covariation, summarizing crown variation into a leaf-economics dimension and two additional biochemical dimensions related to hydro-thermal regulation and defence/metabolism. Mapping these dimensions revealed distinct community-level trait organization alongside substantial species- and crown-level variation within forests. These findings link remotely sensed trait retrieval to environmental filtering and plant functional differentiation, providing a scalable framework for monitoring forest functional diversity.
Southern corn leaf blight (SCLB) is caused by the fungal pathogen Bipolaris maydis (syn. Cochliobolus heterostrophus Drechsler) and is a common disease of fall crops of sweet corn. Phenotyping for SCLB resistance is performed through visual scoring, which is subjective and may limit genetic gain for this quantitative trait. As an alternative, we integrated computer vision (CV)-based phenotyping, genome-wide association studies (GWASs), and predictive breeding approaches to dissect the genetic basis of SCLB resistance. We utilized a sweet corn diversity panel with 693 genotypes, for which whole-genome resequencing produced a high-density single-nucleotide polymorphism (SNP) dataset. Broad-sense heritability for visual scoring ranged from 0.44 to 0.73, while CV-based phenotyping produced estimates ranging from 0.56 to 0.73 in multi-environment resistance trials conducted across 5 years and three locations. We performed GWAS using 16,755,210 SNPs and identified 41 associated SNPs. Genomic selection (GS) models on visual scoring phenotypes achieved moderate prediction accuracies under cross-validation of untested genotypes across characterized environments (0.22-0.47) and high prediction accuracies when predicting tested genotypes in uncharacterized environments (0.49-0.68). Using CV-based phenotypes for GS, we observed prediction accuracies of 0.45-0.47 under the untested genotypes in the characterized environments cross-validation scheme and 0.59-0.62 under the tested genotypes in the uncharacterized environments scheme. GS demonstrated reliability for ranking the individuals across a gradient of environments. These findings identify candidate loci and predictive breeding strategies to accelerate the development of resistant sweet corn cultivars.
Reproduction assets foundThe authors state that all datasets (phenotype data) and analysis code (CV phenotyping script, customized GAPIT script) are publicly available in their GitHub repository, which is listed in allowed_urls.Code · publiche images taken for each plot were saved in JPG format and analyzed using a CV method. Here, we refer to the CV method as a custom Python script written using the OpenCV library version 4.5.0, a set of tools for CV (Bradski, 2000 ). The Python script used for leaf CV image analysis is available in our public GitHub repository ( https://github.com/Resende‐Lab/SCLB‐Disease ).
FIGURE 1
Leaf imaging set up with QR‐coded plot IDs (bottom right) and color checker for computer vision phenotyping of southern corn leaf blight disease severity in sweet corn.
In the CT19 environment, a black cloth attached to a wooden board was used as the background. A wooden frame was used to clamp the leaves down toOpen asset ↗Resende‐Lab/SCLB‐Diseaselines:199-209Dataset · publicBLUP and BayesB model implemented in BGLR.
ACKNOWLEDGMENTS
This work was supported by the National Institute of Food and Agriculture USDA‐NIFA2018‐51181‐28419, USDA‐NIFA2019–05410, and USDA‐NIFA 2022–51181‐38333.
DATA AVAILABILITY STATEMENT
All the datasets and codes used in this study are available in the following repository: https://github.com/Resende‐Lab/SCLB‐Disease
REFERENCES
Amadeu , R. R.
,
Cellon , C.
,
Olmstead , J. W.
,
Garcia , A. A. F.
,
Resende , M. F. R.
, &
Muñoz , P. R.
( 2016 ).Open asset ↗Resende‐Lab/SCLB‐Diseaselines:566-596Plant phenotyping relevance matchOpenAlex · Crossref · checked 14 Sept 2026
Abstract Grapevine leaves have dorsiventral anatomy with distinct adaxial (upper) and abaxial (lower) surfaces. Although morphological descriptor lists and ampelographic literature provide information on both the upper and lower side characteristics, in practice, the color traits of the upper side have become the focus of scientific publications. This study introduces the practical application of the recently developed LeafLaminaMap software and the use of trichromatic color indices in grapevine characterization. We aimed to compare colorimetric information on the adaxial and abaxial leaf surfaces as well as to explore the potential of machine learning models in classification. Five statistical descriptors (mean, standard deviation, contrast, energy, and entropy) were calculated for 25 RGB-based color indices on both leaf surfaces of 120 samples collected from four grapevine cultivars (‘Chardonnay’, ‘Pinot noir’, ‘Sauvignon blanc’, and ‘Syrah’). Data was subjected to multivariate statistical analysis and machine learning classifiers. Results showed that the abaxial leaf surface had stronger cultivar-specific color signatures, supporting its suitability for cultivar discrimination. These findings suggest that RGB-based analysis of both adaxial and abaxial leaf surfaces has potential for grapevine cultivar discrimination, offering a new perspective for cost-efficient plant phenotyping.
Background Understanding the structure of plant seeds cultivated for human consumption and food manufacturing is vital to provide sustainable products as well as to investigate early growth stages. This includes structural variation between different plant species, varieties and cultivars depending on genetic setup, as well as structural modifications upon germination, aging and storing or seed treatment during processing. For plant seeds as multi-component biological materials, structural characterization must extend across multiple length scales, from molecular organization to cellular architecture. Results We apply scanning Small- and Wide-Angle X-ray Scattering (SWAXS) and X-ray Fluorescence (XRF) on yellow pea seeds to combine local structural information on the molecular scale with imaging of cellular structures on the micrometer scale, enabling a comprehensive analysis of hierarchical organization. To identify and characterize heterogeneous regions within the pea seeds, we implement a fitting-free, data-driven segmentation and analysis workflow based on machine learning tools. This approach allows for classification of structurally distinct domains and enables quantitative comparison across samples without relying on predefined models. Furthermore, we incorporate multi-modal analysis by combining structural imaging with complementary elemental information obtained from XRF. The integration of compositional and structural data provides deeper insight into structure-composition relationships. Conclusions This multi-scale, multi-modal approach opens new possibilities for investigating hierarchical structures and their development under diverse conditions and enables systematic comparison between different species or seeds at different developmental stages or exposed to different processing steps. The approach is broadly applicable to various kinds of samples and other hierarchically organized biological materials, which makes it a valuable technique for plant science as well as plant-based food science.
Grain yield is a central target in wheat breeding, yet accurately predicting it remains challenging because it depends on many genes and responds strongly to environmental variation. Genomic selection (GS) has improved breeding efficiency by enabling genome-based prediction of genetic merit, but predictability (PA) for grain yield is often limited under stress environments. At the same time, advances in high-throughput phenotyping (HTP) using unmanned aerial vehicles (UAVs) provide phenomic data that capture environment-responsive plant performance and may complement genomic information. In this study, we evaluated genomic and phenomic models for predicting grain yield in elite bread wheat lines across irrigated, drought, and heat-stress environments. Using a sparse phenotyping framework, we compared parametric and non-parametric models. PA was evaluated within environments and under cross-environment sparse phenotyping scenarios. Genomic models provided a stable baseline and enabled effective information sharing across environments when phenotypic data were incomplete. Phenomics-only models captured environment-specific plant responses but were more sensitive to environmental context. Multiomics models that integrated genomic and phenomic information consistently achieved the highest PA, with the largest gains observed under stress conditions. Overall, our results demonstrate that integrating genomics and UAV-based phenomics within sparse phenotyping designs offers a practical and scalable approach to improve grain yield prediction in wheat.
Reproduction assets foundThe paper's grain yield BLUEs, spectral wavelength BLUEs, and genotypic data are publicly deposited in the CIMMYT data repository (https://doi.org/10.71682/10549399), directly reproducing this paper's phenotyping measurements. No author analysis code with a public URL is stated; other URLs are generic tools/services.Dataset · publicok.com.
Paolo Vitale, Email: p.vitale@cgiar.org.
DATA AVAILABILITY STATEMENT
The datasets generated and analyzed during this study, including best linear unbiased estimates (BLUEs) for grain yield and spectral wavelengths, as well as the corresponding genotypic information, are publicly available in the CIMMYT data repository ( https://doi.org/10.71682/10549399 ).
REFERENCES
Araus, J. L.
,
Kefauver, S. C.
,
Zaman‐Allah, M.
,
Olsen, M. S.
, &
Cairns, J. E.
(2018). Translating high‐throughput phenotyping into genetic gain. Trends in Plant Science, 23(5), 451–466. 10.1016/j.tplants.2018.02.001
Brault, C.
,
Lazerges, J.
,
Doligez, A.
,
Thomas, M.
,
Ecarnot, M.
,
Roumet, P.
,
Bertrand, Y.Open asset ↗CIMMYT data repository · 10.71682/10549399lines:280-433Plant phenotyping relevance matchEurope PMC · checked 14 Sept 2026
Whitefly-transmitted begomoviruses cause tomato leaf curl disease (ToLCD). In India, at least 15 begomoviruses are known to cause ToLCD, posing a major challenge to resistance breeding. Although several Ty resistance loci have been introgressed from wild tomato relatives, variable resistance responses are frequently observed, likely due to mixed infections and the absence of a standardized disease scoring system. Moreover, limited knowledge of the infecting begomoviruses in resistant genotypes has hindered the effective use of donor lines in breeding programs. This study evaluated 17 homozygous Ty-gene donor tomato genotypes under natural epiphytotic conditions and identified the associated begomoviruses. To quantify disease severity, a robust disease scoring system was developed using coefficient of infection (CI) by integrating symptom parameters- leaf curling, leaf smalling, stunting, and fruiting, with population-level disease incidence. Field evaluations for two years revealed that genotypes carrying both Ty-2 and Ty-3 loci showed higher resistance, though variability existed among them. Genotypes with Ty-3 alone or Ty-5 + Ty-6 combinations also displayed substantial tolerance, and five genotypes were identified as highly resistant. Molecular indexing revealed frequent mixed infections and identified multiple begomoviruses, including a newly characterized species, tomato leaf curl Ty Pusa virus, alongside tomato leaf curl New Delhi virus, tomato leaf curl Palampur virus, tomato leaf curl Gujarat virus, and tomato leaf curl Joydebpur virus. These findings highlight a shift in begomovirus predominance and possible recombination-driven emergence of new variants. This study provides an integrated framework for evaluating ToLCD resistance and emphasizes the need for continuous reassessment of resistance sources to ensure durable tomato cultivar development.
Alexander Bucksch · Yong S. Chung · Jennifer Clarke · Stephan Gerth · Philipp von Gillhaußen · Wei Guo · Jana Kholová · Shree Pariyar · Ethan Pickering · Sindhuja Sankaran · Sahameh Shafiee · Sergio Alan Cervantes-Pérez · Stijn Dhondt · Zhiguo Han · Kabir Hossain · William LaVoy · Jonathan P. Lynch · Sonia Negrão · Tony Pridmore (427291) · Hannah Schneider · Stefan Schwartz · Ian Stavness · Shangpeng Sun · Vadez · Lee West · H.J. van de Zedde
Field / plotWhole plant / canopy / plot / field
Plant Phenomics studies the phenotypic dynamics that form plant phenotypes by systematically measuring traits using phenotyping methods from the quantum to the ecosystem levels. It has emerged as an interdisciplinary field advancing sensing, computation, and plant biology. However, Plant Phenomics has lacked a unifying framework that integrates its community's core concepts from the formal and life sciences. Central to this framework is the definition of the phenome as a set of phenes that govern phenotypic dynamics across all spatial and temporal scales of biological and ecological organization and in interaction with the environment. This paradigm moves beyond gene-centric views and recognizes the equal importance of all spatial and temporal scales in forming plant phenotypes, advancing Plant Phenomics as a data-driven discipline and its emerging profession, the plant phenomicist.
Year-to-year climate variability poses a challenge for agriculture by increasing crop yield variability; therefore, there is a need to identify genotypes that can withstand these fluctuations. With the right selection criteria, genotypes with yield stability across variable environmental conditions can be selected. Methods such as Finlay-Wilkinson random regression (FWRR) may allow us to use sparse datasets-common in plant breeding pipelines-and incorporate genomic data to leverage phenotypic information from related genotypes to predict yield stability. Our objective was to examine how the number of environments and the variance among those environments affect stability predictions. We also integrate FWRR as a genomic prediction tool for characterizing yield stability, comparing it to the traditional genomic prediction models as a reference. We used three datasets: one highly unbalanced dataset for oats (Avena sativa L.) and two completely balanced datasets with different numbers of environments for barley (Hordeum vulgare L.) and wheat (Triticum aestivum L.). We fit standard Finlay-Wilkinson (FW) and FWRR models to estimate grain yield and stability under various scenarios. We found that the estimated stability values obtained were similar using balanced datasets for FW or FWRR. FWRR also achieved moderate predictive ability for stability using unbalanced datasets under 10-fold cross-validation (CV1) with new genotypes. In terms of environmental representation, selecting the right set of environments for inclusion in the model was more important than adding more environments. Our results suggest the possibility of using FWRR to select stable genotypes earlier in line development, as well as to design resource-efficient stability-testing schemes.
Salt stress can markedly alter seedling architecture, creating a need for non-destructive three-dimensional (3D) phenotyping methods capable of resolving fine plant structures. However, organ-level segmentation of plant point clouds remains challenging because of leaf overlap, slender stems, ambiguous stem–leaf boundaries, and severe class imbalance. In this study, we developed PTV-SegCo, a task-adapted Point Transformer model for organ segmentation and structural phenotyping of coriander seedlings under salt stress. PTV-SegCo integrates efficient channel attention, gated shallow–deep feature fusion, and a combined cross-entropy–Dice loss to improve representation of fine and minority organ structures. The dataset comprised 60 manually annotated 3D point-cloud samples from 12 cultivation trays repeatedly observed over five acquisition dates under six NaCl concentrations (0, 50, 100, 150, 200, and 250 mmol L −1 ). Because the earliest acquisition represented a particularly challenging developmental stage, these 12 samples were used as a fixed early-stage model-selection set, while samples from the remaining four dates were organized into four date-blocked training–validation configurations. Under this internal model-development protocol, PTV-SegCo achieved mean mAcc and mIoU values of 93.05% and 89.05%, respectively, and showed numerically higher performance than its direct backbone PTV-Seg50. These values should be interpreted as internal comparative results rather than as an unbiased estimate of generalization to unseen cultivation trays, and the present results should not be interpreted as establishing the broad competitiveness of PTV-SegCo against other point-based, convolution-based, graph-based, transformer-based, or plant-specific segmentation architectures. After semantic segmentation, reconstructed scenes were metrically calibrated using the known cultivation-tray dimensions, followed by individual-plant separation and quality control. Four reconstruction-derived structural descriptors—plant height, projected area, voxel occupancy volume, and leaf point ratio—were extracted to characterize temporal structural variation under different NaCl treatments. For treatment-level inference, individual-plant measurements were aggregated within each cultivation tray at each acquisition time, with the cultivation tray treated as the independent experimental unit. Independent manual validation showed close agreement for plant height and projected area, with R 2 values of 0.9969 and 0.986, respectively. Overall, the proposed workflow provides a feasible approach for organ-level segmentation and automated 3D structural analysis of small coriander seedlings under salt stress. The extracted descriptors primarily represent reconstruction-derived spatial characteristics and should not be interpreted as direct indicators of physiological status; voxel occupancy volume and leaf point ratio remain without direct external validation.
ABSTRACT Water-use efficiency (WUE), the ratio of accumulated plant biomass to water lost through transpiration has conventionally been determined using a destructive single-point measurement. Recent advances in high-throughput phenotyping now enable repeated, non-destructive estimation of biomass and WUE. However, these digital measurements must be statistically validated against conventional destructive methods to validate their use as reliable proxies. Therefore, we compared digital biomass determined point clouds produced from multispectral camera scanners with destructive harvests across eight harvests using Samsun tobacco grown under both drought and high-water conditions. WUE efficiency, calculated using the digital biomass estimated from a point cloud and gravimetric water use determinations, were compared to destructive harvest determinations. The coefficient of variation (CV) showed there were no significant differences in digital and destructive measurements for either biomass or WUE. Indicating that digital measurements can be used in place of destructive measurements. Drought plants used significantly less water and were significantly smaller than high-water plants from Harvests 4 through 8. However, there were no significant differences in the ratio of evapotranspiration to leaf area or WUE, indicating that drought plants were simply smaller and used less water than the high-water plants. This work validates that estimating plant biomass from a digital point coupled with continuous gravimetric determination of water use provides a reliable nondestructive measure of WUE in high-throughput measurements across the full plant life cycle. PLAIN LANGUAGE SUMMARY We grew tobacco plants under either a drought or high-water treatment and harvested a portion of the plants every few days for a total of eight harvests. Throughout the experiment, we collected 3D images of the plants and continuously measured pot weight to track plant growth and water use across different developmental stages. Destructive biomass served as the gold-standard measurement. We then compared biomass and water-use estimates generated from the digital measurements with the destructive measurements. The digital approach provided accurate estimates of plant biomass and water use while requiring little hands-on labor and no plant destruction. These nondestructive methods could help plant breeders identify water-efficient plants earlier in the breeding process, accelerating the development of crops that use water more efficiently.
Abstract Tree crowns are complex, three‐dimensional structures whose morphology varies among species, individuals and environments. Although light detection and ranging (LiDAR) provides high‐resolution, single‐tree point clouds that advance species discrimination and the assessment of intraspecific variation in situ, crown shape is still commonly reduced to low‐dimensional metrics (e.g. crown diameter or crown base height), losing much of its three‐dimensional geometric complexity. We introduce a fully 3D geometric morphometric framework that captures crown shape directly from LiDAR point clouds at both species and individual levels. Pre‐segmented LiDAR single‐tree point clouds of eight temperate forest species were converted into three‐dimensional shape representations using radial bounding volumes (RBVs), which partitioned each crown into a standardized set of vertical layers and radial sectors. Surface points automatically digitized from each RBV formed geospatially aligned, 3D pseudolandmark configurations representing geometric morphometric crown shapes. These configurations served as the input data for multivariate analyses of crown shape variation within and between species. Twelve structural traits, including crown and stem dimensions, were extracted from the same RBVs and integrated into analyses of trait–shape associations. The morphospace of crown shape was structured along different axes of variation in broadleaf species than in conifers. Within these groups, species pairs—such as Fagus versus Quercus and Picea versus Pinus —exhibited contrasting intraspecific morphological gradients, with different structural traits driving shape variation in each. Crown base height and total crown height emerged as the strongest predictors of crown shape. Differences in crown shape among species were primarily captured by symmetric components, with asymmetry providing a negligible signal. Interspecific differentiation was largely driven by architectural variation rather than pure size differences. Morphological differences derived from pseudolandmarks and convolutional neural network features exhibited stronger correlations in conifers than in broadleaf species. We present a reproducible, LiDAR‐native framework for quantifying and comparing 3D crown morphology within and across species. Using the RBV approach, geospatially aligned pseudolandmarks can be derived from any pre‐segmented, single‐tree LiDAR point cloud, enabling scalable, multi‐regional analyses of intraspecific variability. This framework provides a robust foundation for integrating crown shape into ecological, evolutionary, silvicultural and modelling studies, including assessments of environmental effects and architectural constraints.
Maize ear geometry (length, width, curvature, and volume) is closely tied to yield and grain-filling outcomes, but existing high-throughput phenotyping pipelines remain constrained by the cost, labor, and specialized hardware they require. We developed and validated a low-cost pipeline that reconstructs a watertight 3-D mesh of a maize ear from a single 20-second video captured with a consumer-grade DSLR on a motorized turntable under uniform LED illumination. Camera poses from a multi-seed COLMAP procedure initialize a Neural Radiance Field (NeRF), and a cylindrical holder of known diameter, visible in every frame, provides automatic metric scaling with downstream geometric quality control. Applied to 300 ears spanning a diverse maize inbred panel, 250 (83.3%) passed automated processing and quality control. Skeleton length agreed with manual caliper measurements across all 250 ears (R^2 = 0.964, RMSE = 4.68 mm), and convex-hull volume agreed with water-displacement volume on a 15-ear subset spanning the full size range (R^2 = 0.982, RMSE = 5.26 mL). Residual length error grew with ear curvature, whereas bounding-box height, which records the same straight-line chord as calipers, showed no such trend; the discrepancy therefore originates in the measurement definition, since calipers record the chord while skeleton length traces the geodesic arc. The capture hardware costs approximately 607 USD, and operator involvement fell from roughly five minutes to one minute per ear, with all downstream processing running unattended. The platform provides a foundation for breeding-scale 3-D ear phenotyping.
To overcome YOLOv11’s limitations in complex field environments, this paper proposed SDD-YOLOv11n, a lightweight real-time detector for soybean diseases. The model reconstructed the backbone using GhostConv to minimize redundancy and integrates a C3k2_Star module to enhance small lesion detection against background noise. Additionally, a Detect Efficient (DE) head further compressed the architecture. Experimental results verified the model's efficiency, achieving a parameter count of 1.88 M and a weight size of 3.9 MB—reductions of 27.3% and 25% compared to YOLOv11n, respectively. Furthermore, the model maintained high detection performance with a Mean Average Precision (mAP50) of 75.6% and an F1-score of 69.3%, demonstrating its effectiveness in balancing architectural efficiency and accuracy in complex field environments.
Surface phenotyping underpins plant science, preclinical animal research and entomology, yet across all three the measurement is almost always a photograph, which records a projection and not the surface itself. Here we present the Gentschinator3000 , an open structured-light platform that brings high-end metric surface measurement within reach of laboratories with no optics expertise, combining documented open hardware, open reconstruction software and analysis workflows for under 4000 Euro in components. It resolves a planar reference to 45 µm local flatness, registers full rotations to a loop closure of 156 µm, and performs stably across acquisition ranges that we define. Applying one workflow to a leaf before and after desiccation, to murine anatomy and to a spread lepidopteran, we find that projection underestimates surface area by 11 to 41 %. That error grows with the condition under study, with the evaluation scale and with the direction of view, so it can confound phenotype comparisons dramatically. In murine limbs a 15-degree change of viewing direction shifts a projected inter-segment angle by up to 23.2 degrees, while the three-dimensional angle does not move. Projection geometry can therefore contribute as much to a measured phenotype as the biology it is meant to quantify.
Reproduction assets foundThe paper explicitly deposits three public Zenodo records: reconstructed 3D surfaces of all specimens (including the leaf and hop cone phenotyping measurements), the authors' analysis notebooks with derived and per-panel source data, and the reconstruction software with build documentation and working examples. All areDataset · publicData availability
1149
The reconstructed surfaces supporting this study are available at Zenodo under
1150
https://doi.org/10.5281/zenodo.22167250.54
1151
Source data for all graph panels are provided with this paper; for panels showing
1152
rendered surfaces, the underlying reconstructions are in the same record.
1153
1154
Code availability
1155
The analysis notebooks, environment specifications, derived data and per-panel
1156
source data are available at Zenodo under
1Open asset ↗Zenodo · 10.5281/zenodo.22167250pdf-raw-page:35 lines:1-52Code · public1151
Source data for all graph panels are provided with this paper; for panels showing
1152
rendered surfaces, the underlying reconstructions are in the same record.
1153
1154
Code availability
1155
The analysis notebooks, environment specifications, derived data and per-panel
1156
source data are available at Zenodo under
1157
https://doi.org/10.5281/zenodo.22167598.55
1158
The reconstruction software, build documentation and minimal working examples
1159
are available at https://doi.org/10.5281/zenodo.22167471.56
1160
The software and analysis notebooks are released under the MIT licence and the
1161
hardware design files under CERN-OHL-P v2. The visible-light platform described
1162
hereOpen asset ↗Zenodo · 10.5281/zenodo.22167598pdf-raw-page:35 lines:1-52Plant phenotyping relevance matchCrossref · checked 14 Sept 2026
Published31 Aug 2026International Journal of Advances in Intelligent InformaticsCited by 0 · OpenAlex ↗
Anton Anton · Supriadi Rustad · Guruh Fajar Shidik · Abdul Syukur
Pepper / chilliPotatoTomatoLeafClassificationVisualization / data managementDisease symptoms / severity
Early detection of plant leaf diseases is critical for minimizing crop losses and supporting precision agriculture. While Convolutional Neural Networks (CNNs) have demonstrated high accuracy in image-based diagnosis, conventional architectures may not optimally balance spatial localization and channel-wise feature refinement, particularly in multi-crop classification settings. This study proposes a redundancy-aware dual-attention architecture, termed ATSA-DenseNet, which integrates the spatial branch of the Convolutional Block Attention Module (CBAM-Spatial) with Efficient Channel Attention (ECA) within a DenseNet121 backbone. Unlike prior dual-attention frameworks that retain full CBAM and introduce channel-level redundancy, the proposed design isolates complementary spatial and channel mechanisms to improve representational efficiency without increasing computational complexity. The framework is evaluated on controlled multi-crop PlantVillage-derived datasets comprising tomato, potato, pepper, and maize. Across both 3-crop and 4-crop configuration, ATSA-DenseNet consistently outperforms baseline DenseNet121 and single-attention variants, achieving 99.94% accuracy and 0.9994 macro-F1 on the 4-crop setting while maintaining a lightweight footprint (6.96M parameters, 2.87G FLOPs). Grad-CAM visualizations indicate improved localization of disease-relevant regions compared to the baseline. While results are obtained under controlled imaging conditions, the findings demonstrate that redundancy-aware dual-attention enhances feature discrimination efficiency in multi-class agricultural classification tasks. Future work will extend validation to real-field datasets with natural variability.
Rice false smut is a major panicle disease that affects rice yield and grain quality and is an important target for resistance evaluation in breeding programs. Accurate field phenotyping is important for disease assessment and resistance screening, yet current assessment relies heavily on manual visual scoring and smut ball counting, which are laborious and subject to evaluator variation. In close-range single-panicle images, false smut balls are often small, dense, occluded, adhered, making automatic detection and severity grading difficult. To address these challenges, we developed a YOLOv12-RSLW detector by integrating RepGhost, SimAM, LSCD, and WIoU into YOLOv12. Detection boxes were then used to guide the Segment Anything Model for panicle and lesion mask extraction, allowing calculation of the lesion-to-panicle area ratio as a supplementary indicator for count-based severity grading. A total of 1911 original field images were collected. After augmentation, the dataset contained 5663 images, including 4531 training images, 566 validation images, and 566 test images. Detection performance was evaluated on the test set, while SAM segmentation was assessed using 80 manually annotated original images. YOLOv12-RSLW achieved 92.06% mAP@0.5, 91.46% precision, and 87.01% recall, with 3.45 M parameters and 6.0 GFLOPs. Compared with the baseline YOLOv12, mAP@0.5 and recall increased by 3.60 and 4.37 percentage points, respectively. Within the augmented dataset, 41.2% of samples initially assigned to Grade 1 and 28.1% of those assigned to Grade 2 met the area-ratio criteria for potential reassignment to higher grades. The framework provides a quantitative approach to rice false smut severity phenotyping and may support future resistance breeding after further validation.
Plant diseases greatly affect agricultural production, especially in developing countries, where prompt diagnosis can be quite challenging due to the limited availability of experts in real-time. Deep learning techniques for image analysis is gaining popularity and are increasingly considered an alternative to traditional manual inspection of plants. This research presents the evaluation of plant leaf disease detection system based on a convolutional neural network (CNN) optimized with different nature-inspired algorithms. The backbone model is based on the EfficientNet-B0 pretrained on ImageNet. Therefore, transfer learning is used to adapt the model to an updated PlantVillage dataset. Experiments have been conducted with multiple nature-inspired algorithms to improve generalisation and training efficiency of the prediction model. Different data preparation techniques have been carefully applied to the dataset, creating a unified approach to ensure consistency in the preprocessing pipeline for the training, validation, and testing phases. Our experiments indicate that application of the grey wolf optimizer (GWO) for tuning key hyperparameters of the model, including dropout, learning rates, and weight decay produced the best results, with an accuracy around 99.45%.
Abstract - Plant diseases can significantly affect plant growth, productivity, and overall health. This research presents a mobile-based plant analysis system designed to identify plants, assess their health condition, and analyze visible diseases from plant images. The proposed system allows users to capture an image using a mobile camera or upload an existing image. The image is processed and analyzed using machine learning and deep learning techniques. A Convolutional Neural Network (CNN) can be used to learn visual features such as leaf shape, color, spots, and disease symptoms for plant identification and disease analysis. The system also provides a health assessment and disease severity indication to support users in understanding the condition of a plant. A Flutter-based mobile application provides the user interface, while Python and Flask can be used for image-processing and model-serving tasks. The proposed approach aims to provide a simple and accessible tool for preliminary plant identification, health assessment, and disease analysis. Key Words: plant identification, plant health assessment, disease analysis, CNN, deep learning, Flutter.
Goal. To substantiate the methodical approaches to analyzing the state of soybean crops based on the results of aerial photography with UAV by comparing manual vectorization, controlled classification according to the algorithm of maximum similarity, and uncontrolled classification of K-Means, as well as to determine the feasibility of their combination with expert visual interpretation to assess the spatial structure of the vegetation cover. Methods. Aerial photography of the test proving ground was performed with the help of the unmanned aerial vehicle DJI Phantom 4 Advanced with the subsequent photogrammetric study of materials and the formation of a highly detailed orthophotoplane. In the QGIS environment, visual decryption and manual vectorization of the main objects of the agrolandscape were carried out with the creation of polygonal layers. For automated mapping, methods of controlled classification according to the algorithm of maximum similarity and uncontrolled classification based on the K-Means algorithm were used. The accuracy of the results was evaluated by comparing the data of automated classifications with the data of manual digitization, which was used as a reference (control) method. On the basis of the results obtained, empirical data were summarized to justify practical recommendations for the application of the studied approaches. Results. The study was conducted on the territory of the research farm of the Separate subdivision of the National University of Life and Environmental Sciences of Ukraine «Berezhany Agrotechnical Institute» (vil. Pavliv, Ternopil district, Ternopil oblast) (49.452057°N; 24,805818°E) in may – august 2025. The obtained cartographic materials made it possible to quantify the areas of the main objects of the agro-landscape and identify problem areas with sparse shoots. Methods of controlled classification showed greater compliance with the digitization data (average deviation — 6.7%) compared to uncontrolled (14.7%), which were effective from the point of view of preliminary assessment of spectrally homogeneous sections, but did not provide an accurate division of shoots by density. Analysis of the spatial structure of coverage made it possible to plan local agrotechnical measures and assess the potential yield. Conclusions. Methodical approaches to analyzing the state of soybean crops based on the results of aerial photography with a UAV equipped with RGB cameras are promising and economically feasible. Automated classification methods are effective for highlighting hard and contrasting objects and small-contoured areas, while a detailed assessment of the structure of the vegetation cover is advisable to carry out using a controlled classification in combination with expert visual interpretation.
All living organisms rely on the movement of ions across cell membranes as the fundamental physical basis of their internal energy and signaling, and plants are no exception. Plants perceive, integrate, and respond to environmental stimuli through electrical signals, classified as action, variation, and system potentials, that are coupled with calcium waves, reactive oxygen species, and hydraulic and hormonal changes to coordinate whole-organism responses despite the absence of a nervous system. Yet most studies characterize these signals using a single feature, such as amplitude or spike duration, in a single tissue, an approach that cannot establish how such signals correspond to the underlying ionic activity, mobility, and structural complexity of the signaling environment, or how this correspondence varies across organs. Here, we correlate plant bioelectrical signals with potential ionic energy flow using a multi-domain framework, combining discrete spike events, continuous waveform properties, spectral composition, and signal complexity applied to leaf, stem, and root recordings from tomato ( Solanum lycopersicum ) exposed to different stimulus. Electrical activity with increased stimulus strength, likely reflecting increased ionic flow, with the root showing the largest response. This suggests plant electrical signaling works as a distributed, ion-based information system, useful for stress monitoring and bio-inspired sensor design.
Plant stress biology has traditionally relied on the analysis of snapshot measurements—including hormone levels, reactive oxygen species, transcripts, and physiological traits—which primarily characterize the current state of the cell, whereas the physical consequences of previous stress exposure remain considerably less accessible to direct measurement. This distinction may be particularly important under natural conditions, where plants experience recurrent, sequential, and combined stresses. This raises a fundamental question: can a cell, after physiological recovery, retain a measurable residual physical state that reflects aspects of its previous stress history and influences its response to subsequent stress? Here, we propose Environmental Phase Imprinting (EPI) as a testable biophysical hypothesis according to which environmental stress may leave a measurable imprint on the physical state of biomolecular condensates that persists after cessation of the initial exposure. EPI is not proposed as a new form of biological information or as an established mechanism of stress memory, but rather as a potential physical substrate, correlate, or consequence of previously described forms of cellular stress memory. To operationalize this hypothesis, we introduce the Plant Condensate Code (PCC), a multidimensional conceptual framework designed to move from a static “snapshot” of cellular state toward a time-resolved physical trajectory. PCC integrates complementary characteristics of condensate populations, including morphology, dynamics, molecular mobility, material state, and molecular composition, across a sequence of states encompassing baseline, stress, adaptation, recovery, and the post-stress state.We propose that the trajectory of condensate states, rather than any single measurement, may contain information about cellular stress history and may help explain differences in responses to recurrent or combined stress. Multimodal approaches, including live-cell imaging, fluorescence recovery after photobleaching (FRAP), molecular mobility analysis, microrheology, Brillouin microscopy, quantitative phase imaging, and molecular profiling, could provide complementary measurements of this physical state. PCC is further positioned within our broader conceptual research program encompassing Cytoplasmic Phase Homeostasis, Cytoplasmic Phase Sensing, and the Plant Threat Matrix (PTM)—a proposed six-state framework for describing plant physiological states under stress. Within this framework, physical measurements of condensate and cytoplasmic states may represent one possible approach for defining and quantitatively characterizing cellular physiological states. Finally, we discuss the potential application of this conceptual framework to stress-resilience phenotyping, evaluation of biostimulants, and selection for stress tolerance, while clearly distinguishing experimentally established phenomena from hypotheses and conceptual proposals. The proposed framework may provide a foundation for the development of a new direction in biophysical phenotyping of plant stress resilience, complementing molecular and physiological approaches to the study of stress memory.
This paper presents an integrated system design for autonomous quadcopter flight path generation using the MAVLink protocol and a custom Ground Control Station (GCS) for precision agricultural crop monitoring. The system combines three coverage path algorithms (Boustrophedon, Spiral, and Energy-Optimized), a Pixhawk 4 / ArduPilot flight stack, a MicaSense RedEdge-P multispectral payload, and a ROS2-based GCS for mission planning, telemetry, and vegetation-index-based crop health assessment. The 2.8 kg quadcopter (450 mm frame, 4-cell LiPo) achieves 22–25 minutes of flight time. Across five field sizes (0.5–10 ha), the Energy-Optimized path achieved 96.5% coverage efficiency with 4.2% overlap and a 12.4% energy reduction over the Boustrophedon baseline. NDVI-based crop segmentation achieved pixel accuracy of 92.5% (maize), 94.1% (rice), and 90.8% (wheat), and four-class crop-health classification achieved a weighted F1-score of 90.0%. MAVLink 2.0 command latency averaged 15.8 ms with 99.3% packet delivery at ranges up to 800 m. An ablation study showed additional gains of 1.5–3.1% coverage from wind compensation and 2.1–2.8% from terrain-following.
Cauliflower emergence rate and seedling growth are key indicators of field conditions and varietal potential. Traditional manual surveys are unsuitable for continuous monitoring across multiple varieties. This study integrates UAV RGB imagery with the DualSlim-YOLO model to estimate cauliflower emergence rates and monitor seedling growth. Built on YOLOv11, the model incorporates a lightweight feature extraction structure and an optimized detection-scale configuration. It reduces computational complexity while maintaining detection accuracy, thereby improving the efficiency of cauliflower seedling detection. DualSlim-YOLO achieved P, R, F1-score, mAP@0.5, and mAP@0.5:0.95 of 95.35%, 96.75%, 96.05%, 98.55%, and 86.65%, respectively. The number of parameters was reduced by 38.61%, while the inference speed increased by 22.16%, demonstrating good lightweight performance. Based on this model, UAV images of 171 cauliflower varieties acquired at 7, 21, and 28 d after transplanting were used for seedling detection and emergence rate estimation. In addition, 18 time-series seedling phenotypic traits were extracted, enabling a comprehensive quantitative evaluation of emergence dynamics and early-growth performance across multiple cauliflower varieties. This method effectively screens cauliflower varieties for high emergence rates, rapid emergence, and excellent seedling growth performance. It provides technical support for high-throughput, nondestructive seedling phenotyping and early germplasm screening under field conditions.
Visible-near infrared (Vis-NIR) spectroscopy provides rapid crop disease assessment; however, poor model generalizability remains a major limitation when models developed for a specific period are applied to batches collected at different times, primarily due to variations in physicochemical properties such as chlorophyll content, moisture level, surface texture, and tissue structure, which induce shifts in spectral distributions across batches. This study investigates deep domain adaptation to enhance cross-batch transferability for sugarcane disease classification. Two batches of healthy and symptomatic leaves were collected at different times using the same spectrometer. A customized one-dimensional convolutional neural network (1D-CNN) was trained on Batch 1 and adapted to Batch 2 using labelled samples through two strategies: retraining only the fully connected layers or fine-tuning all network parameters. Both strategies achieved 94% accuracy, with precision 0.92, sensitivity 0.97 and specificity 0.98, outperforming the non-adapted model and standard-free calibration transfer methods, namely Correlation Alignment and Transfer Component Analysis. These findings demonstrate that deep domain adaptation substantially improves the robustness and transferability of Vis-NIR classification models across heterogeneous sampling batches.
Sebastián Varela · Jeremy Ruhter · Erik J. Sacks · Xuying Zheng · Dylan Allen · Anna Hale · Cory Landry · Xianyan Kuang · B. J. Long · Ernst Cebert · Yan Zhu · Shatabdi D. Proma · Sehijpreet Kaur · Diego Jarquin · Jesse Morrison · Andrew D. B. Leakey
Abstract The integration of digital technologies for high-throughput field phenotyping is critical for accelerating crop improvement in agriculture. However, extracting traits from remote sensing data remains constrained by fragmented workflows, manual intervention, and limited interoperability among existing tools, resulting in delays that hinder timely biological insight and decision-making. To address these challenges, we present PhenoStream (Phenotyping Streaming), a scalable, end-to-end cyberinfrastructure designed to automate the full lifecycle of aerial imagery–based phenotyping, from data acquisition to plot-and genotype-level inference. The framework integrates automated data ingestion from distributed field sites, geospatial processing, and AI-enabled trait extraction within a unified, user-accessible graphical interface. Its modular and extensible architecture supports adaptable trait modeling and seamless integration of new data sources, enabling deployment across diverse crops, environments, and experimental designs. We demonstrate the system across a large multi-location field trial network of bioenergy crops, where it enables high-throughput characterization of spatiotemporal growth dynamics, genotype-by-environment (G×E) interactions, and predictive modeling of key agronomic traits. By significantly reducing processing latency and manual effort, the platform facilitates near-real-time analysis and reproducible workflows. This work establishes a generalizable and scalable pathway for operationalizing very-high-spatial resolution aerial phenotyping in agricultural research. By bridging data acquisition and analytics, the end-to-end cyberinfrastructure provides a foundation for integrating heterogeneous and unstructured data streams—including remote sensing, environmental, and management data—toward data-driven decision making in agriculture.
Background: Faba bean is an important legume crop valued for its nutritional and soil-enriching benefits, yet its productivity is severely affected by foliar diseases. Automated image-based detection using deep learning provides a rapid and reliable approach for early disease identification and improved crop management. Methods: This study developed a machine learning-based automated framework for multi-class classification of Faba bean leaf diseases using transfer learning with the VGG16 convolutional neural network. A dataset of 8,021 RGB images collected under natural field conditions was used, comprising four classes: healthy, rust, gall and chocolate spot. Images were resized to 224 × 224 pixels and normalized prior to training. The pretrained convolutional layers of VGG16 were frozen and a custom classification head with global average pooling and dropout regularization was added. Model performance was evaluated using classification metrics. Result: The proposed model achieved an overall classification accuracy of 92.34% and a macro-averaged F1-score of 0.9227 on the test dataset. Strong classification performance was observed across all disease categories, with particularly high predictive accuracy for healthy and rust classes. The findings demonstrate the effectiveness of transfer learning for plant disease detection and highlight its potential for scalable, automated crop health monitoring in precision agriculture.
Xylella fastidiosa (Xf) is among the most devastating phytosanitary threats to Mediterranean agriculture, causing Olive Quick Decline Syndrome (OQDS). Since containment efficacy depends on timely intervention, scalable in-field screening tools are needed. This study evaluates a portable digital electronic nose, based on a carbon-nanotube sensor array combined with artificial intelligence algorithms, for the in-field screening of Xf through volatile organic compound (VOC) profiling. Three replicate acquisitions were performed on 130 olive trees (390 measurements) across four cultivars (Cellina di Nardò, Ogliarola Salentina, Ogliarola Barese, and Leccino) in four Italian regions (Apulia, Calabria, Lazio, and Tuscany). Plant status was assigned from the official status of the sampling area (demarcated OQDS focus versus Xf-free area) and supported by real-time quantitative PCR (qPCR) on every plant; within demarcated sites, plants with undetectable DNA in sampled twigs were retained as Xf+ following phytosanitary criteria, giving 216 infected and 174 healthy samples. The multidimensional sensor signals were processed with an optimized Shallow Neural Network. Under plant-grouped 80/20 validation, keeping each plant’s replicates in the same subset, the model achieved (93.3 ± 4.0)% accuracy, (97.7 ± 3.5)% sensitivity, and (88.2 ± 8.3)% specificity (mean ± SD). A feature-importance analysis revealed a reproducible, though not chemically resolved, VOC-related response pattern. This low-cost, portable Internet of Things (IoT) device offers a proof-of-concept screening approach for Xf surveillance, pending plant-level and external validation.
Abstract Purpose Evaluate how soil and canopy sensing can map within-block variability in tart cherry orchards and identify indicators robust enough for repeatable management decisions. Methods Soil apparent electrical conductivity (ECa) was mapped in spring 2022 across four commercial tart cherry blocks (8.5–10.5 ha; approximately 3,500 trees per block), followed by canopy sensing in 2023–2024. Canopy structure was measured using unmanned aerial vehicle (UAV) photogrammetry and mobile terrestrial laser scanning (MTLS) using light detection and ranging (LiDAR), and canopy density using mobile ceptometry. Spatial layers were aligned to per-tree grid cells. An August 2025 campaign compared UAV- and LiDAR-derived tree height with ground-truthed height. Results Soil-to-canopy relationships were weak to moderate but consistent within blocks ( r = 0.10–0.40), with strength and direction varying by site conditions. Canopy density was more strongly associated with UAV-derived volume than height. UAV-derived 90th-percentile height best predicted ground-truthed height ( R ² = 0.89; RMSE = 0.34 m), whereas LiDAR showed a weaker relationship and greater error ( R ² = 0.70; RMSE = 0.52 m). Cross-sensor agreement was moderate to strong ( r = 0.41–0.65). Per-tree rankings were stable between years for UAV height and volume. Conclusion Whole-block sensing revealed persistent spatial patterns that could support management-zone delineation. UAV photogrammetry provided accurate canopy metrics, MTLS offered measurements suited to routine orchard operations, ceptometry added seasonal canopy-density information, and ECa provided soil context. Occasional ECa mapping combined with strategically timed UAV surveys and other sensors as needed could reduce redundant sensing while supporting fertilizer evaluation, pruning, and labor allocation.
Rice leaf diseases pose a significant threat to global food security by reducing crop productivity and causing substantial economic losses. The traditional diagnosis method is manual method, which is low in efficiency, subjective and not suitable for large-scale agricultural monitoring. Despite the advances in automated disease detection using deep learning methods like CNNs, GANs, and transfer learning models, these techniques remain highly computational, not very flexible, and struggle to perform well in different imaging scenarios. Considering these drawbacks, this paper introduces an Explainable Deep Q-Learning based CNN framework which employs CNN-based feature extraction and Deep Q-learning-based adaptive policy optimization for rice leaf disease classification. The proposed model continuously refines the classification actions through reward-based learning, which makes the model more robust in various agricultural imaging environments, in contrast to traditional supervised CNN models that have static classification decisions. The proposed model achieved 98.5% accuracy, 98.52% precision, 98.50% recall, and a 98.51% F1-score, outperforming existing CNN, GAN, reinforcement learning, and transformer-based methods. It also offers a high computational efficiency of 14.2 GFLOPs, 248 MB memory consumption, ~ 48 min of training time, and 6.8 ms inference time per image suitable for resource constrained applications in agriculture. The results demonstrate the effectiveness, scalability, and practical applicability of the proposed framework. The proposed framework performs well on benchmark datasets but more research in the deployment of the edge-devices under different real-world agricultural settings will be investigated in future work.
Reproduction assets foundThe paper trains its Deep Q-CNN rice leaf disease classifier on public Kaggle rice leaf image datasets, which are cited with explicit public URLs and qualify as paper-specific phenotyping image inputs. The authors' own derived data/analysis artifacts are only available upon request, so no authors' code or trained modelDataset · publicSoni Gautam. Rice Leaf Bacterial and Fungal Disease Dataset. Kaggle. Available:Open asset ↗Kagglepdf-page:24 lines:1-94Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Plant diseases are still posing a challenge to the productivity, quality of crops, and food security, especially in locations where field diagnosis is based on manual visual inspec-tion. This paper assesses deep learning network-based automated classification of plant leaf diseases on public RGB leaf-image datasets, such as the Kaggle New Plant Diseases Dataset (Augmented) and PlantVillage images. They investigated four archi-tectures: EfficientNetV2B0, ResNet152V2, DenseNet201, and one hybrid Vision Trans-former (ViT)-based model. The steps of the experiment involved loading the dataset, exploratory analysis, preprocessing, resizing, normalizing, augmentation, transfer learning, independent model training, and evaluation metrics such as accuracy, preci-sion, recall, F1-score, training curves, testing results, and confusion matrices. The hy-brid ViT-based model was reported to have the best accuracy of 99.5%. On the smaller seven class subset of PlantVillage, EfficientNetV2B0 scored 98.11%. On the 38-class dataset, DenseNet201 improved test accuracy (97.34) and validation classification ac-curacy (around 98). ResNet152V2 scored 97.01 on the 38-class test set. The results demonstrate that CNN and transformer-based models can help to recognize plant diseases accurately whereas hybrid attention-based structures provide a promising path to enhance fine-grained classification. Since the model notebooks had varying class settings and splits, the comparison is seen as a model-structured assessment as opposed to a precisely identical benchmark across all architectures.
Reproduction assets foundThe paper's phenotyping inputs are two public plant leaf-image datasets explicitly named in the Data Availability statement: the Kaggle New Plant Diseases Dataset (Augmented) and the PlantVillage dataset, both with public URLs. No author code, models, or supplementary materials are deposited (supplementary materials: 'Dataset · publicy available. Plant leaf images were
obtained from the Kaggle New Plant Diseases Dataset (Augmented) and PlantVillage
datasets. The datasets contain publicly accessible RGB images of healthy and diseased
plant leaves used for supervised image classification research.
DATASET SOURCES
Kaggle New Plant Diseases Dataset (Augmented):
https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset
PlantVillage Dataset: https://plantvillage.psu.edu/All processed data, experimental configurations, and model implementation details are
described within the manuscript. Additional materials may be made available from the
corresponding author upon reasonable request.
ACKNOWLEDGMENTS
The author acknowOpen asset ↗Kaggle · new-plant-diseases-datasetpdf-raw-page:24 lines:1-23Dataset · publicgmented) and PlantVillage
datasets. The datasets contain publicly accessible RGB images of healthy and diseased
plant leaves used for supervised image classification research.
DATASET SOURCES
Kaggle New Plant Diseases Dataset (Augmented):
https://www.kaggle.com/datasets/vipoooool/new-plant-diseases-dataset
PlantVillage Dataset: https://plantvillage.psu.edu/All processed data, experimental configurations, and model implementation details are
described within the manuscript. Additional materials may be made available from the
corresponding author upon reasonable request.
ACKNOWLEDGMENTS
The author acknowledges Istanbul Aydin University for academic support and research
guidance duringOpen asset ↗PlantVillagepdf-raw-page:24 lines:1-23Code / dataset availability confirmedCrossref · checked 15 Sept 2026
Precision agriculture is becoming more and more of a challenge that requires the use of intelligent systems that are able to predict stress and prevent yield loss before it is too late. Traditional methods of agricultural surveillance are predominantly reactive with irrigation demands being based on thresholds or individual yield forecasts models that do not represent the intricate spatio-temporal interactions that exist between crop physiology, soil status, and environmental stresses. Besides, the majority of the current practices do not have an autonomous decision-making approach to preventive intervention which leads to inefficient use of water and slows down the response to stress. This paper suggests a cognitive UAV-assisted agro-surveillance system to predict yield vulnerability caused by crop stress and optimize adaptive irrigation with the help of spatio-temporal deep and reinforcement learning. The framework combines UAV-obtained RGB and multispectral and thermal imagery with measurements of soil sensors and meteorological data obtained with the Crop Health and Environmental Stress Dataset. A new GeoSpatio-TRiNet model is used to acquire long-range spatial relationship, time stress development, and diffusion of stresses across agricultural regions. The model predicts the vulnerability trajectories of the stress instead of the direct yield regression, and this allows early detection of yield risk. Such predictions serve to generate a cognitive environmental state of a Soft ActorCritic (SAC) reinforcement learning agent that autonomously computes zone-based irrigation behaviors to reduce the recurrence of stress at the minimum water usage cost. As shown by the results of the experiment, the proposed framework has a stress forecasting accuracy of 96.3% and performs much better than the traditional machine learning, CNN-based, and transformer-based baselines. The system also decreases the predicted yield vulnerability by 46.6 and enhances water-use efficiency by 41.1 as compared to irrigation strategies based on rules. The results confirm the usefulness of spatio-temporal intelligence with predictive control in terms of effectiveness, and the proposed framework is a scalable and sustainable solution to precision agriculture of the next generation.
Reproduction assets foundThe paper uses the public Kaggle Crop Health and Environmental Stress Dataset (UAV RGB/multispectral/thermal imagery plus soil/weather measurements and stress labels) as its phenotyping data source, and the authors provide an explicit public GitHub repository for the analysis code.Dataset · publicThe current research is based on the Crop Health and Environmental Stress Dataset, which is a publicly available
dataset on Kaggle, specially created to help perform a spatio-temporal analysis of crop health in response to changing
environmental and water-stress factors [26].Open asset ↗pdf-raw-page:10 lines:1-62Code · publicturn: Final zone-wise stress predictions 𝐶
𝑡
𝑧, Yield vulnerability trajectories 𝑉𝑡
𝑧, Optimal adaptive irrigation policy
𝜋∗
End Algorithm
Code availability:
The data used to support the findings of this study are included in the article.
Code availability:
The code used in this research work is available in the following link.
https://github.com/replyvenugopal/Cognitive-UAV-Driven-Agro-Surveillance
4. Result and Discussion
The architectural agro-surveillance solution, which is proposed to be executed by UAVs, is executed through a
modular and scalable software framework to guarantee reproducibility and extensibility. The experiments are all
performed in Python as a main programming languageOpen asset ↗github.com/replyvenugopal/Cognitive-UAV-Driven-Agro-Surveillancepdf-raw-page:24 lines:1-55Code / dataset availability confirmedOpenAlex · arXiv · checked 5 Sept 2026
High-throughput phenotyping depends on accurate 3D reconstruction of plants across growth stages, yet the development and evaluation of temporal completion methods are limited by the lack of datasets with complete geometric ground truth. To address this challenge, we introduce SynthCrop4D, a procedurally generated synthetic dataset of temporally evolving plant point clouds that provides controllable noise, occlusion, and complete plant geometry for benchmarking reconstruction methods. Using this dataset, we evaluate a two-stage pipeline that combines spatial denoising and temporal point cloud completion. First, a denoising module removes structural artifacts from raw laser-scanned point clouds. The resulting data are then processed by an Adaptive Temporal PoinTr model that reconstructs the current growth stage (t) using information from the previous stage (t-1), enabling recovery of regions missing due to self-occlusion. We evaluate the proposed framework on both SynthCrop4D and the real-world Pheno4D dataset (tomato and maize) under settings with and without denoising. Results show that denoising substantially improves reconstruction quality, with the best configuration achieving a Chamfer Distance of 0.0061 on SynthCrop4D (Temporal PoinTr + Mamba-DG) and an F-Score of 0.2080 on Pheno4D (Vanilla PoinTr + Mamba-DG). We further demonstrate the use of completed point clouds for phenotypic trait extraction, including plant height, canopy width, and convex hull volume, obtaining hull-volume MAEs of 0.021 on synthetic data and 0.343 on real data. Together, SynthCrop4D and the proposed pipeline provide a benchmark and methodology for temporal plant reconstruction and high-throughput crop phenotyping.
Reproduction assets foundThe paper's authors explicitly state that source code, implementation details, and pre-trained model weights are publicly available on GitHub, and that the paper-specific SynthCrop4D synthetic dataset can be reproduced via scripts in that codebase. Pheno4D is a cited prior public dataset, not a paper-specific asset.Code · publicThe source code, implementation details, and pre-trained model weights for this study are publicly available on GitHub at https://github.com/Mrudul2006/3d_plant-reconstruction .Open asset ↗Mrudul2006/3d_plant-reconstructionlines:1724-1761Plant phenotyping relevance matchEurope PMC · OpenAlex · checked 15 Sept 2026
Crossman SG, Wang M, Nowotarski SH, McClain ML, Sankari S.
Alfalfa / lucerneX-ray / CTRoot2D/3D reconstructionVisualization / data managementGrowth / development / phenology
The symbiotic relationship between the legume Medicago sativa and the soil bacteria Sinorhizobium meliloti results in the formation of nitrogen-fixing root nodules. Traditional destructive methods, including paraffin sectioning, vibratome sectioning, and cryosectioning, have been applied to visualize how bacteria occupy the nodule, making it extremely difficult to obtain reliable three-dimensional information. These approaches are often combined with fluorescent labeling or staining, which can introduce additional stress affecting plant growth and nodule formation. MicroCT has emerged as a relatively quick, easy, and robust tool for plant biology that can non-destructively visualize plant histological features in three dimensions (3D), thereby avoiding destructive artifacts during sample preparation and ensuring high-fidelity 3D reconstruction. While microCT has been applied to legume root nodules, a detailed established protocol that documents the process from plant harvest and sample preparation to scanning and software visualization is lacking. In this study, we show a step-by-step microCT workflow using Medicago sativa as a model. The protocol includes nodule excision from roots, fixation, contrast enhancement, mounting, scanning, and three-dimensional reconstruction. Critical parameters affecting elements such as image quality, tissue preservation, and contrast are highlighted. Using this approach, it is possible to visualize the overall tissue organization, bacteroid-infected cells, and vascular bundles in three dimensions without physically sectioning the nodules. The pipeline described here provides a reproducible method for non-destructive, high-resolution imaging of native root nodules and is likely adaptable to other legume species, offering researchers a practical tool for studying nodule structure and bacterial organization within nodules in 3D.
Abstract Non-invasive, high-throughput phenotyping tools are needed that can identify environmental effects on plant structure and function to diagnose factors responsible for reduced growth in commercial and non-commercial settings. In this study, we explored whether the integration of 3D-multispectral (3D) and 2D-hyperspectral imaging (HSI), aided by machine learning (ML), could be used to identify environmental stress treatments imposed during plant growth. Controlled environment-grown Nicotiana Benthamiana plants were subjected to a range of abiotic treatments – including different growth irradiances, heat treatment and drought stress – with the treatments resulting in differences in shoot height, biomass, leaf area and spectral reflectance. ML models were trained to identify these treatments using morphological and spectral traits measured at 27, 29, 31, and 34 days after sowing (DAS). A 3D-multispectral scanner was used to obtain information on plant height, biomass, and leaf area. A visible and near-infrared (VNIR) HSI camera provided detailed spectral information for deriving spectral indices including the Normalised Difference Vegetation Index (NDVI), Photochemical Reflectance Index (PRI) and Normalized Difference Red Edge (NDRE). Manual measurements provided baseline comparative data. The 3D-multispectral scanner reliably estimated above-ground traits, with high correlations between manual and scanner-derived measurements. The ML models accurately differentiated among environmental stress treatments, with the fused 3D+HSI model achieving the best overall predictive performance across all evaluated metrics compared with models based on either imaging modality alone. Results demonstrated the effectiveness of combining 3D-multispectral and 2D-HSI data with ML analyses for non-destructive, high-throughput phenotyping. The integration of these techniques enabled non-destructive, high-throughput identification of environmental stress treatments imposed during plant growth.
Existing conversational plant-phenotyping platforms are difficult for plant scientists to use and lack the reliability scientific research demands: failed analyses are reported as valid measurements rather than flagged as missing, statistical tests run without checking assumptions, predictions carry no uncertainty estimate, and specialised hardware limits accessibility. We present PhenoIntel, a lifecycle-aligned multi-agent web platform that turns the full machine-learning workflow into a reliable, user-friendly phenotyping system. Nine specialised agents divide the analysis into stages, from image collection through model selection, inference, and reporting, rather than handing the whole task to one AI manager. Independent checks separate these stages, and every agent reads from and writes to one shared, fixed-structure record, so an inconsistent output from one stage is caught before it reaches the next. Uncertainty is matched to each model family, conformal prediction, detection-confidence spread, or Monte Carlo Dropout, rather than applied uniformly, and quality thresholds adapt to crop and task instead of one global cutoff. When no suitable model exists, PhenoIntel can propose, validate, and integrate a new one on its own. The model repository spans ten trained models across five crops and four imaging modalities. Classification models reach Macro F1 of 0.78-0.996; object-detection models reach 0.96 mAP@50 with a 54% reduction in counting error over an unoptimised baseline; and a temporal model reaches held-out Macro F1 of 0.7050. PhenoIntel runs in a browser on standard hardware, requiring no GPU, and a 1,200-test automated suite confirms complete pipeline execution. Every result carries calibrated uncertainty, validated statistics, and FAIR-compliant provenance, a combination existing conversational phenotyping tools do not offer.
Reproduction assets found論文固有の解析コードとモデル資産を公開するGitHubリポジトリを本文中の根拠とともに確認しました。Code · publiccode, model checkpoints, and the 1,200-test automated suite referenced
throughout this paper are maintained in a version-controlled
repository, available at
https://github.com/Naren1704/PhenoIntel-InternshipOpen asset ↗Naren1704/PhenoIntel-Internshiplines:2047-2163Plant phenotyping relevance matchCrossref · checked 15 Sept 2026
Agricultural production faces significant annual losses due to plant diseases, with economic impacts exceeding 40 million dollars and contributing to acute hunger affecting over 281.6 million people in 2023. The timely and accurate identification of plant diseases through leaf image analysis is crucial to mitigate these losses and ensure global food security. This study proposes a lightweight Convolutional Neural Network model, inspired by the MobileNet architecture, designed to classify various plant leaf diseases efficiently. Leveraging a publicly available dataset, this research focuses on developing a model that balances high performance with computational efficiency, making it suitable for real-world applications in resource-constrained environments. The proposed model, named LDPNet, achieved an outstanding accuracy of 99.62%, alongside precision, recall, F1-Score, and AUC metrics of 99.16%, 99.08%, 99.11%, and 99.99%, respectively. A comprehensive comparative analysis was conducted against MobileNetV2 and a reference model from previous research, highlighting the superior performance of LDPNet in terms of both accuracy and efficiency. The results demonstrate that the proposed architecture not only maintains high classification performance but also significantly reduces the number of parameters, making it a practical and scalable solution for plant disease identification. This study contributes to the growing field of agricultural technology by providing a robust, lightweight, and efficient tool for early disease detection, with the potential to enhance crop management and reduce economic losses in agriculture.
Abstract Background and aims Plant volatile organic compounds (VOCs) change dynamically with plant development and in response to environmental conditions. However, their potential as non-invasive indicators of phenological progression remains poorly explored. In this study, we developed a framework integrating automated VOC sampling, time-resolved VOC profiling, and machine-learning analysis for the non-invasive assessment of plant phenology. Using soybean ( Glycine max (L.) Merr.), we investigated whether development-associated temporal variation in VOC emissions could delineate and predict developmental phases. Methods We collected VOCs daily under controlled environmental conditions from 16 to 43 days after sowing, spanning the transition from vegetative to reproductive stages, using an automated sampling system coupled with thermal desorption-gas chromatograph-mass spectrometer (TD- GC-MS). To characterise temporal changes in VOC profiles associated with phenological progression, we analysed the daily VOC data using a multi-step pipeline combining statistical filtering and similarity-based network analysis. We defined VOC-derived developmental phases from similarity patterns in the VOC profiles, then developed and evaluated machine-learning models to predict these phases. Key results Seven VOCs exhibited distinct phase-dependent dynamics, including green leaf volatiles and monoterpenes showing characteristic temporal changes during phenological progression. Network-based clustering of VOC profiles resolved five developmental phases closely aligned with conventional developmental stages. A machine-learning model predicted these phases from the VOC profiles with high predictive accuracy on independent test data, demonstrating that phenological progression could be quantitatively inferred from VOC emission patterns. Conclusions Our findings support VOC profiling as a reliable and non-invasive approach for assessing phenological progression in soybean. By extracting temporally structured VOC signals, this framework captures developmental information that may be difficult to obtain through visual observation alone, particularly after canopy closure. VOC profiling offers a practical tool for monitoring crop developmental dynamics and has broader potential for plant phenotyping and precision crop management.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Supplement · publicThe peak area matrix obtained from the MS- DIAL analysis (Supplementary Dataset S1) was filtered to remove unreliable features.Open asset ↗lines:66-69Plant phenotyping relevance matchOpenAlex · checked 5 Sept 2026
Introduction Leaf shape is a genetically determined crop phenotype, and its accurate classification underpins soybean germplasm assessment and genetic improvement. Manual classification is highly subjective and struggles to distinguish morphologically similar leaves, while mainstream supervised classification demands large labeled datasets and incurs high development costs. Efficient feature frameworks for soybean leaf categorization are still insufficient. Methods In this study, 581 biologically replicated terminal leaflets sampled from 194 soybean varieties were analyzed at the single-leaflet level using traditional morphological indices and novel leaf contour angular features. Unsupervised K-means clustering was used to classify soybean leaflet morphological phenotypes; t-SNE was applied exclusively for dimensional reduction visualization, while Welch's ANOVA combined with Games-Howell post-hoc tests was adopted to detect inter-cluster phenotypic differences. Clustering stability and external consistency against manual visual labeling were further quantified via Adjusted Rand Index to comprehensively verify the reliability of grouping outputs. Results The results revealed no significant difference in leaflet edge complexity (p = 0.41) between two manually divided leaf groups distinguished by overall leaf outline similarity; these two morphologically similar leaf clusters failed to be fully separated even though the first two principal components accounted for 90.2% of total variance. For K-means clustering, k = 3 achieved better overall performance with a Calinski–Harabasz (CH) index of 395.55, Davies–Bouldin (DB) index of 1.03, and silhouette coefficient (SC) of 0.38, compared with k = 4. Nevertheless, the angular feature attained an F-value of 951.62 in driving sample reallocation across clusters, serving as the core indicator for fine subdivision at k = 4. Under k = 4 clustering, all six morphological indices differed significantly among the four groups (p < 0.05). Additionally, the number of cross-clustered samples increased from 66 to 119 as k rose from 3 to 4, with 96.6% of cross-clustering attributed to the leaflet contour angular feature. Discussion This research provides a novel reference and technical support for the automated identification and fine classification of soybean leaf morphology.
Yaning Zhao · Guodong Zhang · Jiaying Rong · Mingyue Gao · Peipei Dang · Tianjun Zhang · Hao Suo · Zhijun Wang · Ziyong Cheng · Panlai Li · Jun Lin
Raman / spectroscopyObject detectionPhysiological trait estimationWater status / transpiration
ABSTRACT Real‐time, accurate water monitoring is a crucial technical foundation for industrial, environmental, and biological research. However, traditional detection methods typically include destructive processes and suffer from response delays. Near‐infrared (NIR) luminescent metal halides offer a novel solution to this challenge, but they still face issues such as ultraviolet excitation and low photoelectric conversion efficiency. Herein, a luminescent material was synthesized based on the blue‐light‐excited lead‐free perovskite Cs 2 HfCl 6 :Te 4+ /Mo 4+ , in which energy transfer (ET) from Te 4+ to Mo 4+ enables highly efficient NIR luminescence in the 800–1200 nm wavelength range. Upon encapsulation with a commercial blue light chip, the fabricated NIR light‐emitting diode device achieved a photoelectric conversion efficiency of up to 16.1%. By utilizing the absorption characteristics of water molecules in the NIR spectrum and receiving signals via a sensor, an interactive learning process based on a neural network machine learning algorithm was employed, achieving an estimation accuracy of up to 98.6% for plant water content. This non‐destructive and precise NIR detection module provides a new solution for the real‐time monitoring of crop physiological status and holds broad application prospects in the fields of precision agriculture and plant science.
The increasing world population necessitates new sustainable nutrient sources, making microalgae like Chlorella sorokiniana interesting due to its rich nutrient profile and sustainable cultivation methods. With genetic optimization tools like CRISPR/Cas9, microalgae as a nutrient source can be improved even further. However, degradation of the rigid cell wall of microalgae, and thereby developing protoplasts, is often necessary prior to transformation, but monitoring protoplast development in spherical, single-celled organisms like C. sorokiniana is challenging using bright-field microscopy. Carbotrace 480 and 630 were tested as fluorescent markers of the cell wall of a C. sorokiniana mutant for protoplast detection, and Carbotrace 480 was successfully used to distinguish protoplast from normal cells in a cell suspension. The enzymes Driselase, Glucanex, Snailase, and Saczyme were tested in different combinations to degrade the cell wall of the mutant, with Snailase as the most effective yielding ~60 % protoplasts. This study provides a quick and easy tool for monitoring protoplast development in the microalgae C. sorokiniana, the first step to improve C. sorokiniana as a sustainable nutrient source using genetic optimization tools like CRISPR/Cas9.
Main conclusion We established a protocol for reliable nuclear visualization in Charophyceae, revealed diverse nuclear organization across cell types and species, and identified suitable cells for genome size estimation via flow cytometry. Charophyceae are multicellular green algae closely related to land plants and are established model systems for understanding plant evolution. Yet key cellular parameters like genome size remain poorly characterized. We combined fluorescence microscopy, transmission electron microscopy (TEM), and flow cytometry to characterize nuclear diversity across cell types and species of Characeae and to identify a cell type suitable for genome size estimation. Among three DNA-intercalating fluorochromes, propidium iodide labeled nuclei most reliably. Nuclear morphology varied widely across cell types: mononucleated cells were found in vegetative apical cells, the coronula of oogonia, and spermatogenous filaments of antheridia, whereas multinucleation predominated in other tissues, e.g., cortical cells, spine cells, stipulodes and rhizoids. Nuclei in Chara hispida showed a significant gradient in cross-sectional area along the thallus axis. In the apical internodes, nuclei were larger and more heterogeneous, whereas in the basal internodes they were smaller and more uniform, which is consistent with possible endopolyploidy. TEM confirmed the nuclear identity of crescent-shaped structures. Relatively large nuclei were found in rhizoids and spine cells. Only Sphaerochara canadensis showed an organized nuclei pattern. Whole-thallus preparations did not yield a defined nuclear peak by flow cytometry, but antheridia of Chara tomentosa produced a sharp peak, from which a genome size of 5.32 pg (1C) was estimated. The protocol established here provides a simple, reproducible framework for visualizing nuclei and estimating genome size in Charophyceae, and helps address longstanding questions in this group, such as the mechanisms and functions of multinucleation, the site of meiosis, and genome evolution.
Faba bean yield reflects complex relationships among genotype, environment, and agronomic traits. This study evaluated an explainable Tabular Prior-data Fitted Network (TabPFN) framework for estimating plot-mean single-plant yield and prioritizing traits using 398 plot-level observations, 13 measured agronomic predictors, and six derived features. On the reference 80/20 split, TabPFN achieved the best values for all four test metrics (R2 = 0.8746, RMSE = 1.9132 g plant−1, MAE = 1.0819 g plant−1, and MAPE = 8.16%). The Friedman test detected differences among the six models (χ2(5) = 16.75, p = 0.005); Nemenyi comparisons distinguished TabPFN from HistGradientBoosting and SVR, whereas the Holm-corrected Wilcoxon analysis confirmed only the TabPFN–SVR difference. Across 10 repeated 80/20 splits, TabPFN obtained the highest mean test R2 (0.8614 ± 0.0691), ranked first in eight splits, and produced a higher R2 than every tuned baseline in at least eight splits. SHAP, permutation importance, and LOCO analyses emphasized pod-, seed-, and biomass-related predictors. Repeated-split ablation showed that derived features improved TabPFN consistently, whereas removing selected target-proximal yield variables reduced performance for every model. The framework is therefore a harvest-time trait-estimation and trait-prioritization tool rather than an early-season forecasting system. Notably, TabPFN achieved this performance without the 100-trial Optuna search used for each baseline; only n_estimators was screened over four prespecified values.
Barley ( Hordeum vulgare L.) is a major cereal crop whose soluble dietary fiber (SDF) offers significant health benefits, yet conventional SDF determination methods are time-consuming, labor-intensive, and destructive to samples. This study developed a rapid, non-destructive method for SDF quantification in barley using mid-infrared (MIR) spectroscopy combined with a Boruta-based partial least squares (PLS) hybrid approach. A total of 280 barley grain samples were subjected to Fourier-transform infrared (FTIR) spectral acquisition from 2000 to 650 cm -1 , with reference SDF values determined by the association of official analytical chemists (AOAC) 991.43 enzymatic-gravimetric method. The Boruta algorithm selected 165 informative wavenumbers out of 363, reducing the variable space by 54.5%. The developed Boruta-PLS model achieved excellent predictive performance with a coefficient of determination for the training set ( R 2 C ) of 0.9695 and for the test set ( R 2 P ) of 0.9601 and a root mean squared error for the training set (RMSE C ) of 0.7464%, and for the test set (RMSE P ) of 0.7340%, substantially outperforming the full-spectrum PLS model with an R 2 P of 0.7759 and an RMSE P of 1.7387%, as well as conventional wavelength selection methods including variable importance in projection (VIP), competitive adaptive reweighted sampling (CARS), and uninformative variable elimination (UVE). The selected wavenumbers were predominantly located in chemically relevant regions at 1000-1200 cm -1 and 1500-1700 cm -1 , confirming model interpretability. This Boruta-PLS approach provides a rapid, non-destructive, and cost-effective alternative for SDF quantification, with strong potential for high-throughput screening and real-time quality monitoring of barley samples.
In the process of agricultural intelligence, precise detection of plant organs serves as the foundation for core tasks such as crop phenotyping analysis and yield prediction. However, in complex field environments, small targets such as citrus flowers and shoots face challenges including scale variation, background interference, and dense occlusion, which severely impact detection accuracy. This study improves the YOLOv10 model by introducing the BAM (Bottleneck Attention Module) attention mechanism and GIoU (Generalized Intersection over Union) loss function, constructing a YOLOv10s-BAM-GIoU model suitable for citrus flower, fruit, and shoot recognition. The BAM attention mechanism enhances the model's feature extraction capability for small target organs under complex backgrounds through parallel channel and spatial attention branches; the GIoU loss function improves the localization accuracy of densely occluded targets by optimizing the geometric alignment between predicted and ground-truth boxes. Validation experiments were conducted on a self-constructed dataset. The experimental results show that the improved YOLOv10s achieves significant advantages in comprehensive detection accuracy, with an mAP50 of 89.1%, representing an improvement of 2.9%~9.5% over the original YOLOv10s and other comparative models. In fine-grained category detection, the model achieves mAP50 of 91.2%, 83.6%, and 92.5% for shoots, flowers, and fruits, respectively. Furthermore, while maintaining high detection accuracy, the model achieves a detection speed of 23.6 ms per frame, meeting real-time detection requirements. The research results demonstrate that the improved YOLOv10s model integrating the BAM attention mechanism and GIoU loss function achieves an optimal balance between accuracy and speed in citrus organ detection tasks, providing a preferred solution for field real-time detection systems.
Abstract Purpose Rapid and non-destructive detection of pigment and nutrient traits in tomato ( Solanum lycopersicum L.) leaves is essential for precision fertilization and greenhouse management. However, most existing studies focus on individual traits (e.g., chlorophyll or nitrogen) with isolated models, limiting the establishment of robust analytical workflows across growth stages and cultivation conditions. This study systematically evaluated hyperspectral imaging workflows for estimating pigment and nutrient traits in tomato leaves. Methods Hyperspectral images were collected at the stages of flowering-fruiting, ripening and harvest from tomato plants supplied with nitrogen at 0, 210, 300 and 390 kg N ha⁻¹. Total contents of chlorophyll, total nitrogen and nitrate were measured by standard biochemical assays for model calibration and validation. Result After sample partition with four strategies, seven spectral preprocessing methods were evaluated, namely moving average (MA), Savitzky-Golay smoothing (SG), Gaussian filtering (GF), median filtering (MF), normalization, baseline correction and standard normal variate (SNV), with normalization, MA and SNV yielded the best predictive performance for chlorophyll, total nitrogen and nitrate, respectively. For feature wavelength selection, competitive adaptive reweighted sampling (CARS) and successive projections algorithm (SPA) were used with CARS yielding the best performance for total chlorophyll and nitrate prediction, while SPA was optimal for total nitrogen prediction. By application of the above optimal methods, random forest (RF), support vector machine (SVM), eXtreme Gradient Boosting (XGBoost) and convolutional neural network (CNN) models were developed to predict pigment and nutrient indicators. Conclusion The SVM performed best for chlorophyll ( R c ²=0.823, R p ²=0.431) prediction, while the CNN achieved higher accuracy for total nitrogen ( R c ²=0.826, R p ²=0.780) and nitrate ( R c ²=0.851, R p ²=0.753) prediction. Overall, leaf nitrogen-related traits were predicted more reliably than total chlorophyll, for which validation performance remained limited. Impact These findings demonstrate that sample partitioning, spectral preprocessing, wavelength selection and model selection should be optimized for each target trait rather than applied uniformly. This study provides a methodological basis for non-destructive assessment of tomato leaf N status and precision fertilization management.
While most phenotyping platforms rely primarily on image-based measurements, advanced plant characterization requires the integration of active physiological sensing modali- ties such as chlorophyll fluorescence. We present an autonomous robotic platform designed to perform targeted fluorescence measurements on plant leaves. The system combines 3D plant reconstruction, geometric analysis, and motion planning to localize suitable measurement points and generate collision-free trajectories for a robotic manipulator. A dense 3D model of the plant is reconstructed from multi-view data and used to extract candidate leaf surfaces based on orientation, accessibility, and sensing constraints. These targets are then integrated into a task-level planning framework that guides the end-effector to precise contact or near-contact configurations required for point-based fluorescence acquisition. The platform enables automated, repeatable, and spatially resolved physiological measurements that go beyond passive imaging. By tightly coupling perception, geometric reasoning, and manipulation, the proposed system provides a robotics-driven approach to high-resolution plant phenotyping and opens new directions for autonomous agricultural inspection and plant-aware manipulation.
Reproduction assets foundThe paper states its code is publicly available in the authors' SonyCSLParis GitHub repository (Plant3DImager), which implements the phenotyping perception and motion-planning pipeline. The exact full URL is split across a line break in the supplied text, so the verifiable allowed URL prefix is used.Code · public2 The code is available at https://github.com/SonyCSLParis/ 3 See for example at https://www.youtube.com/watch?v=Open asset ↗SonyCSLParis/pdf-page:4 lines:1-61Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Abstract Technological advances have expanded the adoption of digital technologies in agriculture, helping to reduce labour effort, increase profitability, improve crop efficiency and productivity, enhance product quality, mitigate environmental impacts, and promote human health. This context also extends to soybean farming, a sector of major economic importance in Brazil. Most importantly, Brazil has the global leadership in soybean production with biological nitrogen fixation (BNF) replacing chemical fertilisers. The research and evaluation of BNF is limited by manual counting of nodules, a time-consuming procedure. This study presents SoyNodules, designed for the automatic identification of soybean nodules, consisting of a dataset of images. The dataset includes 1,701 images acquired under controlled conditions: 1,662 images of soybean roots with nodules and 39 images of isolated nodules without roots. A total of 49,210 nodule instances are manually annotated with bounding boxes. SoyNodules was designed to promote reuse and interoperability in alignment with the FAIR principles (Findable, Accessible, Interoperable, Reusable) and to support the development, training, and evaluation of computer vision and deep learning methods for precision agriculture.
Reproduction assets foundThe paper is a data descriptor for SoyNodules, an annotated dataset of 1,701 soybean root/nodule images with 49,210 bounding-box annotations, publicly deposited on Zenodo with a DOI. The same repository also hosts the authors' annotation-format conversion script (AnyLabeling to Pascal VOC/COCO), per the Code AvailabilDataset · publicThe SoyNodules dataset, released as version 1.0, is publicly available on Zenodo [28]
at https://doi.org/10.5281/zenodo.22081914.Open asset ↗Zenodo · 10.5281/zenodo.22081914pdf-page:9 lines:1-43Plant phenotyping relevance matchOpenAlex · checked 11 Sept 2026
Accurate quantification of rice aboveground biomass (AGB) is critical for crop monitoring but remains challenging due to the complex nonlinearity arising from the coupling of plant density, spatial structure, and internal dry matter distribution. To address the limitations of single-source remote sensing, this study proposed a Three-Dimensional Dry Matter Distribution Integration (3D-DMI) model, which establishes a physically interpretable framework decomposing AGB into dry matter density ( ρ ), horizontal projection distribution ( S ), and vertical cumulative distribution ( h d ) components. Guided by this framework, a core subset of six features (Red_650, MTCI, G_correlation, R_correlation, LPI, and HPA0_99) was extracted from UAV-based multispectral, RGB, and LiDAR data using a dual-step feature selection approach combining Maximum Information Coefficient (MIC) and Distance Correlation (dCor). A Random Forest (RF) regression model was then developed to estimate AGB across the entire growth season. The results demonstrated that the 3D-DMI model achieved excellent performance with an R 2 of 0.920, an RMSE of 0.184 kg/m², and an RPD of 3.544, significantly outperforming any single-sensor approach. Single-feature analysis revealed that while LiDAR-derived structural features provided the fundamental basis for biomass estimation, they encountered inherent saturation bottlenecks during late growth stages. Feature contribution analysis based on SHAP further quantified that LiDAR-derived features dominated the estimation process (68.5% contribution), providing the volumetric basis, whereas RGB textures (18.3%) and multispectral features (13.3%) provided indispensable supplements. Ultimately, this study established a robust, physically grounded computational paradigm for high-precision UAV-based rice biomass monitoring across the entire growth cycle.
Sebastian Varela · Jeremy Ruhter · Erik Sacks · Xu-ying Zheng · Dylan Allen · Anna Hale · Cory Landry · Xian-yan Kuang · Blake Long · Ernst Cebert · Yan Zhu · Shatabdi D. Proma · Sehijpreet Kaur · Diego Jarquin · Jesse Morrison · Andrew D. B. Leakey
Abstract The integration of digital technologies for high-throughput field phenotyping is critical for accelerating crop improvement in agriculture. However, extracting traits from remote sensing data remains constrained by fragmented workflows, manual intervention, and limited interoperability among existing tools, resulting in delays that hinder timely biological insight and decision-making. To address these challenges, we present PhenoStream (Phenotyping Streaming), a scalable, end-to-end cyberinfrastructure designed to automate the full lifecycle of aerial imagery-based phenotyping, from data acquisition to plot- and genotype-level inference. The framework integrates automated data ingestion from distributed field sites, geospatial processing, and AI-enabled trait extraction within a unified, user-accessible graphical interface. Its modular and extensible architecture supports adaptable trait modeling and seamless integration of new data sources, enabling deployment across diverse crops, environments, and experimental designs. We demonstrate the system across a large multi-location field trial network of bioenergy crops, where it enables high-throughput characterization of spatiotemporal growth dynamics, genotype-by-environment (G×E) interactions, and predictive modeling of key agronomic traits. By significantly reducing processing latency and manual effort, the platform facilitates near-real-time analysis and reproducible workflows. This work establishes a generalizable and scalable pathway for operationalizing very-high-spatial resolution aerial phenotyping in agricultural research. By bridging data acquisition and analytics, the end-to-end cyberinfrastructure provides a foundation for integrating heterogeneous and unstructured data streams-including remote sensing, environmental, and management data - toward data-driven decision making in agriculture.
Chlorophyll fluorescence provides sensitive information on plant photochemical responses, but its measurement requirements can limit high-throughput application. This study investigated whether RGB imagery could approximate chlorophyll-fluorescence-derived photochemical status across garden plant species during progressive soil drying. A Photochemical Status Index (PSI) was constructed by principal component analysis from five highly correlated JIP-test energy-flux variables (RC/CS, ABS/CS, TRo/CS, ET2o/CS, and RE1o/CS). The dataset comprised 50 aggregated species-by-soil-moisture-stage observations representing ten species and five sequential soil-moisture stages. Eleven RGB-derived variables were evaluated, and a partial least-squares regression model was assessed using nested leave-one-species-out validation, with all data-dependent procedures repeated within each outer training fold. PC1 explained 96.5% of the shared variation among the fluorescence-derived fluxes. The predictors g, GLI, ExG, ExGR, and CIVE were retained in all ten outer folds. The final model yielded a pooled out-of-fold R2 of 0.469, an RMSE of 1.585, and an MAE of 1.183. However, species-specific R2 ranged from −0.179 to 0.959, and a calibration slope of 0.509 indicated prediction-range compression. These findings provide proof-of-concept evidence of moderate RGB-based approximation of fluorescence-derived photochemical status, but inconsistent species transferability and the common soil-moisture/time gradient require external validation before practical deployment.
Agricultural operations such as pepper harvesting, fruit counting, and field phenotyping rely on accurate visual recognition and instance segmentation algorithms. However, pepper fruits in complex field environments often exhibit slender and curved shapes, partial occlusion, ambiguous boundaries, and adhesion between adjacent instances. Existing object detection and instance segmentation methods therefore struggle to obtain complete fruit masks, which adversely affects subsequent fruit counting, contour measurement, and picking-point localization. To improve the instance segmentation accuracy of occluded peppers in complex field scenes, this study proposes OccPepSeg-YOLO, an improved model based on YOLO11n-seg. First, a P2FreqFusion module is introduced to fuse shallow, high-resolution detail features with deep semantic features, thereby enhancing the representation of fruit edges and tip regions. Second, an ASC module is designed to model the directional and scale-related morphological characteristics of pepper fruits, while a BoundaryGate module strengthens responses at occlusion interfaces and boundaries between adjacent instances. Finally, an OccPepSegment multi-scale prototype segmentation head is constructed, and a BDoU loss function is introduced to improve the boundary consistency of instance masks. Experiments on a self-constructed field-pepper instance segmentation dataset showed that OccPepSeg-YOLO achieved M-P, M-R, M-mAP50, and M-mAP50–95 values of 93.87%, 92.09%, 97.17%, and 82.31%, respectively, representing improvements of 5.59, 3.18, 3.83, and 9.52 percentage points over YOLO11n-seg. Further comparisons with representative YOLO-based instance segmentation models, including YOLOv8n-seg, YOLOv9c-seg, YOLO12n-seg, and YOLOv26n-seg, demonstrated that OccPepSeg-YOLO achieved the best overall segmentation performance. In particular, its M-mAP50–95 exceeded the best competing result obtained by YOLOv9c-seg by 8.35 percentage points. Under a unified repeated-inference protocol on an RTX 3090 GPU using FP32 precision, a batch size of 1, and 640 × 640 inputs, OccPepSeg-YOLO achieved a mean inference latency of 15.801 ± 1.238 ms, a P95 latency of 17.323 ms, and a throughput of 63.29 FPS. These results demonstrate that the proposed model can produce more complete pepper instance masks under leaf occlusion, fruit overlap, and complex background conditions, providing technical support for field-pepper recognition, fruit counting, and visual perception by agricultural robots.
High-resolution monitoring of forest structure and productivity is essential for effective natural resource management. However, monitoring approaches such as field-based forest inventories or extensive lidar campaigns are costly, time-intensive, and spatially limited. Therefore, inexpensive and accessible methods are needed. SatCHM (Satellite Canopy Height Model) was developed to be an accessible and open-source tool for researchers, allowing for site-specific and temporally flexible predictions of canopy height with limited computational resources. SatCHM requires four inputs: panchromatic satellite imagery, solar and sensor angle metadata of satellite imagery, digital elevation models (DEMs), and lidar-produced CHMs for an area of interest. After SatCHM pre-processes inputs, data is loaded into a collection of convolutional neural networks (CNNs) for image-to-image regression. This ensemble cooperates to yield high-resolution predictions (up to 0.5-meter) of three-dimensional tree structure with discernible tree crowns across a broader defined area of interest. After calculating the mean absolute error for each prediction output, the median of these mean absolute errors was 6.06 meters.
ABSTRACT Leaf shape is a fundamental trait of plant ecological strategies, influencing biotic interactions and ecosystem functioning. However, established quantitative metrics fail to capture subtle variations and irregularities, require user-based reference points or are challenging to compare among taxa with broadly different leaf shapes. In addition, established metrics typically conflate (aggregate) leaf edge complexity and macro-shape complexity, despite their independent functional significance and genetic foundations. Here, we introduce an entropy-based framework to quantify two new complexity metrics: edge complexity and macro-shape complexity. Based on three case studies, we show that these metrics outperform aggregate metrics in predicting Quercus robur chemical traits, provide more intuitive interspecific classifications, and strongly align with human perception. In addition, edge and macro-shape complexity show high complementarity, while aggregate metrics are highly redundant and typically strongly related to leaf area. Emerging as the strongest predictor of leaf chemistry and key visual cue for complexity as perceived by humans, the effects of edge complexity highlight the under-appreciated functional significance of leaf margins. Our framework and the proposed entropy-based complexity metrics thus promise to help unlock the potential of growing digital image archives of leaves, including images from herbaria and fossils, and are technically readily applicable to shapes of algae, bacteria, pollen, and beyond. The accompanying package ShapeComplexity enables the broad application of entropy-based metrics, providing a powerful tool to explore how the shape of organisms and biological structures influences ecological strategies, biotic interactions, and ecosystem functioning while tracking spatial and temporal variation.
Reproduction assets foundThe paper's authors publicly release their ShapeComplexity analysis code (Rust) on GitHub, used to compute the paper's leaf edge- and macro-shape complexity metrics. Supplementary data/analysis code are on Dryad, but that URL is not in the allowed list. RMBG is a generic third-party background-removal model, not a phenCode · publicThe complete, open-source Rust-code (The Rust Team, 2025 ) is publicly available on GitHub ( https://github.com/Thornbach/ShapeComplexity ), ensuring transparency and reproducibilityOpen asset ↗Thornbach/ShapeComplexitylines:86-94Plant phenotyping relevance matchEurope PMC · checked 15 Sept 2026
Emerged Technologies include imparted potentiality to generate adequate food to converge ultimatum of society. However, aspects like, climate change, plant disease, refuse at pollinators as well as others are demanding to farmers. The presence of diseases pulls up the maturation of their corresponding species. Plant disease recognition using image processing and machine learning (ML) with data collected from Internet of Thing (IoT) sensors has been receiving greater amount of interest in recent years, however timely disease detection remains a demanding issues. Despite early disease detection though reduces the risk of destruction to plants nevertheless the noise present in the sensor networks (i.e. Internet of Things) while collecting internet scrapped images compromise overall precision and accuracy. To address on this research gap, i.e., addressing noise with timely disease detection, in this work ML based plant disease detection method called, Intersection Histogram and Rotational Invariant Principal Component Regression (IH-RIPCR) is introduced. IH-RIPCR technique is dividing as pre-processing and feature extraction. First with the raw plant dataset obtained as input from Corn or Maize Leaf Disease Dataset, pre-processing is done employing Intersection Histogram based Contrast Enhancement model. Second with the obtained pre-processed contrast enhanced images is subjected as input to ML-based Rotational Invariance Principal Component Regression feature extraction model to extract relevant features pertaining to corn or maize leaf images for disease detection in an accurate and precise manner. Experiments are conducted with corn or maize plant leaf dataset with different existing methods to verify hypothesis potentiality of IH-RIPCR technique. Significance of technique is evaluated based on numerous parameters in terms of PSNR, processing time, precision, recall and accuracy respectively.
Raj Nirmal Rajendran · M. Timothy Rabanus‐Wallace · Jinghan Lu · Geoffrey B. Fincher · Caterina Selva · Matthew R. Tucker · James Hunt · David Moody · Mohammad Pourkheirandish
Structural failure of cereal stems during late-season climate extremes is a critical determinant of yield stability. In barley, breakage of the stem below the spike, known as head loss, leads to major yield losses, particularly in hot and dry regions where the crop is widely grown. Despite a predicted increase in head loss risk due to global warming, current understanding of the genetic, physiological, anatomical, and environmental factors that control head loss remains limited. Overcoming these knowledge gaps is essential to providing a systems-level strategy for barley breeders to develop climate-ready cultivars that are resilient to stem breakage and suitable for industry adoption. Here, we review present knowledge and highlight opportunities for innovation to mitigate head loss through interdisciplinary approaches that combine precise phenotyping through mechanical testing of stem strength and flexibility, high-throughput phenotyping through drone-based spike counting, and genetic modification strategies informed by studies on hormonal regulation and cell wall composition. Coupled with genotypic data, these efforts will enable the development of a genomic selection platform to facilitate future breeding programs. The framework and tools discussed here are broadly applicable to improving stem resilience in other cereal crops.
Sharma S, Lupo Y, Munoz J, Cochetel N, Nunez V, Gaspar A, Torres-Lomas E, Cantu D, Diaz-Garcia L.
GrapevineMultispectral / hyperspectralRoot
Adventitious root formation is critical for the cost-effective propagation of grapevine rootstocks, and poor rooting limits the adoption of new rootstocks derived from underutilized Vitis L. species. We evaluated 308 accessions representing 18 Vitis species over three growing seasons, scoring rooting at two developmental stages, callus-stage and post-transplant, together with root biomass, cutting weight, and a derived transplant- response index. Phenotypic variation was extensive within and among species, and species rankings depended on the trait considered: V. riparia , V. rupestris and V. californica ranked among the top five species for all four rooting traits, whereas V. arizonica and V. acerifolia rooted well at the callus stage but only intermediately after transplanting. Repeatability was moderate to high for root weight and callusstage rooting and lower for post-transplant rooting. Between-species differences accounted for most of the genetic variance in callus-stage rooting but little of that in cutting weight. After removing differences among species, accessions originating from wild sites with lower dry-season precipitation rooted better and produced more root biomass. Genome-wide association analysis of 3.4 million single-nucleotide polymorphisms identified 54 significant markers resolving into 18 independent loci across four traits. Candidate genes implicate auxin-linked cell proliferation, cell wall and lignin remodeling, and solute transport. Genomic and phenomic prediction achieved moderate accuracies across traits and seasons, including for previously unevaluated accessions, and combining spectral with genotypic data improved performance; accuracy was essentially flat between 5,000 and 50,000 markers. These results provide a framework for broadening the germplasm base of grapevine rootstock breeding. Plain Language Summary Grapevines are almost always grown as two plants joined together: a fruiting variety grafted onto a rootstock that supplies the root system. Nurseries build these plants from dormant cuttings, so a rootstock is only useful in practice if its cuttings root easily. Almost all commercial rootstocks descend from just three wild North American grape species, in part because cuttings of other species are believed to root poorly. We grew cuttings from 308 wild and cultivated grapevines representing 18 species over three years and measured how well each one rooted, first in the callusing room and again after the young plants were transplanted. Rooting ability varied a great deal, and several species outside the usual three rooted as well as the standards. Vines originally collected from places with drier summers tended to root best. We also located regions of the grape genome linked to rooting, and showed that the rooting ability of a vine can be predicted from its DNA or from light reflected by its leaves. Together, these results give breeders a way to screen a much wider range of wild grapevines for rootstock development before testing them in a nursery. Core Ideas Rooting ability varies widely across 18 Vitis species, well beyond the three used in rootstock breeding. Callus-stage and post-transplant rooting behave as genetically distinct stages of root formation. Accessions from collection sites with drier summers rooted better and produced more root biomass. GWAS resolved 18 loci implicating auxin signaling, cell wall remodeling, and solute transport. Genomic and phenomic prediction reached moderate accuracy and was insensitive to marker density.
The intrinsic properties of plants offer numerous opportunities for scientific and technological advancement. Considerable efforts have been directed toward developing plant-on-chip platforms to investigate cellular responses to external stimuli, including chemical, mechanical, and electrical cues. In this study, we present a fluidic platform using polydimethylsiloxane (PDMS) and a printed circuit board (PCB), integrated with electrochemical impedance spectroscopy (EIS) detection. Various experimental conditions were examined, including ionic and pH stimulation, as well as membrane dimensions, with the onion inner membrane treated as a black-box system. The measurement results are presented as Nyquist plots, and a resistance model incorporating multifactorial influences is proposed. Impedance variations in plant cells serve as a basis for electrical modulation. To explore these properties, we converted acoustic signals into electrical inputs and recorded the outputs after being modulated by onion inner epidermal cells. A transfer function analysis was subsequently performed. Our results indicate that the plant cell-on-chip (PCOC) platform holds promise for further investigations into plant cell properties. The impedance results suggest that plant cells can respond to different external stimuli, enabling modulation of the electrical properties. These findings lay the groundwork for future studies on cellular electrical characteristics and the development of preliminary bioelectrical circuits.
Maize diseases affecting leaves, stalks, and ears can substantially reduce yield and quality; therefore, rapid and accurate recognition in complex field environments is important for intelligent agricultural monitoring. To address the large-scale variation, weak fine-grained texture, and strong background interference associated with multi-part maize diseases, this study proposes YOLOv11-MPD (YOLOv11 for Maize Multi-Part Disease Detection), a maize disease detection algorithm based on YOLOv11n. The method jointly improves spatial position awareness, shallow detail preservation, local-context modeling, key semantic-region enhancement, and lightweight detection-head reconstruction. RFCAConv, C3k2_RFCAConv, and Detect_LSDECD are introduced into the baseline network to strengthen directional texture modeling, multi-scale feature aggregation, and detection-head feature representation. FG-RFCAConv, HGD-C3k2, LCA-C3k2, and GRN-BiAttn are further designed for high-frequency differential gated detail compensation, P3 high-resolution detail enhancement, local-context fusion, and global-response-normalized attention regulation, respectively. Experimental results show that YOLOv11-MPD achieves Precision, Recall, mAP50, and mAP50-95 of 72.3%, 72.8%, 79.5%, and 50.2%, improving YOLOv11n by 2.4, 2.5, 2.9, and 2.4 percentage points, respectively, while reducing parameters from 2.6 M to 2.4 M. These results indicate that, within the scope of the dataset used in this study, YOLOv11-MPD improves multi-part maize disease detection under complex field conditions. However, the current conclusions are limited to the constructed dataset, and further validation using larger multi-region, multi-season, and multi-device datasets is required to evaluate its broader generalization ability.
Lodging in sorghum presents a significant challenge for plant breeders due to the trade-off between lodging resistance and grain yield. Manually measuring lodging across thousands of plots is time-consuming, expensive, and error-prone, making selection for lodging resistance challenging in breeding programs. Unmanned aerial vehicle (UAV)-derived metrics provide a potential high-throughput alternative; however, it remains unclear whether photogrammetric heights derived from UAV imagery can estimate plot-level lodging severity in large sorghum breeding trials. This study developed a framework for predicting plot-level lodging from UAV imagery across 2,675 sorghum breeding plots. Multi-temporal canopy height data were collected at two critical time points: maximum crop height and at manual lodging assessment. Height percentiles were extracted from UAV-derived point clouds generated using photogrammetric algorithms. These data were used to develop parametric, non-parametric, and ensemble prediction models, which were evaluated using three statistical metrics. The ensemble model, averaging predictions from all models, achieved the highest accuracy with Pearson correlations of r = 0.80-0.84 and lowest root mean square error (RMSE=16-18%), explaining 64-70% of variation in manual lodging counts. Model diagnostics and iterative refinement, including inspection of UAV imagery and dataset curation, had minimal impact on model performance, demonstrating the robustness of the approach. Model performance was consistent across sites, with minimal effects of stratified sampling on accuracy, confirming the ensemble approach as optimal for plot-level lodging assessment. This study demonstrates that integrated multi-temporal UAV imagery offers a practical alternative to labor-intensive manual evaluation methods by enabling high-throughput lodging assessment suitable for implementation in sorghum breeding programs.
Adam Ziółkowski · Franciszek Błaś · Luiza Tymińska-Czabańska
Field / plotLiDAR / point cloudRootMorphology / geometry measurementRoot system architecture
Coarse root systems govern tree anchorage, yet remain among the least documented components of tree architecture: excavation is irreversible, and established 3D methods rely on specialist scanners and lengthy post-processing. We evaluated whether a consumer smartphone records exposed coarse root architecture metrically, and which traits agree most closely with manual measurement. Four fully exposed Scots pine (Pinus sylvestris L.) root systems in northwestern Poland were scanned with an iPhone 17 Pro running Scaniverse, at about 30 min of acquisition and 5 h of processing per tree. Clouds were registered, cleaned and oriented to magnetic north in CloudCompare; of eight architectural metrics, four were validated against manual references at 95 cross-sections on 44 roots, and four were exploratory. Visible root length (root-mean-square error, RMSE, 22.2 cm, 8.4%), azimuth (RMSE 3.58°, mean absolute error 2.47°) and depth (RMSE 3.18 cm, 14.9%) agreed most closely with the reference; 70 of 77 first-order roots were detected with no false positives. Diameter was the weakest metric and the only one dependent on the operator (RMSE 0.46 and 0.29 cm for two operators on the same clouds). Smartphone LiDAR thus turns an irreversible excavation into a permanent, measurable record of the traits relevant to anchorage, provided that centimetre-level diameters are not required.
Abstract Deep convolutional networks now classify leaf images on curated benchmarks such as PlantVillage with accuracies close to the measurement ceiling of those datasets, which has shifted the open questions away from raw accuracy toward two under-reported issues: which components of a composite pipeline actually cause the result, and whether the components that matter on laboratory images are the same ones that matter on field photographs. We address both with a classification framework evaluated under a protocol that fixes every development decision before the independent test data are read. A Mask R-CNN stage localizes the dominant leaf, the accepted box is expanded by a validation-selected 5% margin and resized to a shared 224 by 224 input, and the same localized image is passed to ResNet50, InceptionV3, and MobileNetV2 together with an extractor that produces eighteen colour and shape descriptors. The forty-five class probabilities and eighteen descriptors form a sixty-three-dimensional input to a neural meta-classifier developed by three repetitions of stratified five-fold cross-validation. On a locked 3,101-image PlantVillage test partition of fifteen classes the framework reached 99.77% accuracy and 99.75% macro F1 with seven misclassifications, and a one-component-at-a-time ablation confirmed that every stage contributed. The same design was then trained and evaluated entirely within a separate thirteen-class PlantDoc field-image dataset, where it reached 90.03% accuracy and 89.40% macro F1. This is a within-PlantDoc experiment and not a controlled-to-field transfer test, so the figure measures how the pipeline behaves on field imagery rather than how a PlantVillage-trained model survives a domain shift. The central finding comes from running the identical ablation on both datasets: on clean images the ensemble breadth and descriptors provide the incremental gains, but under field conditions the ordering changes, and leaf localization and learned fusion become the decisive components. Removing localization cost 2.83 accuracy points and replacing the neural fusion with soft voting cost a further 2.08 points, the two largest effects on PlantDoc. The contribution is a controlled and transparent account of where each component of a localization-aware plant disease classifier earns its place, and of how that ordering shifts between laboratory and field acquisition.
Reproduction assets found保存済みの本文根拠を更新済みルールで再検証し、公開資産1件を確認しました。Dataset · publicThe classification subset used here (20,638 images spanning fifteen pepper-bell, potato, and tomato classes) was obtained from PlantVillage, which is openly accessible at https://www.kaggle.com/datasets/emmarex/plantdisease.Open asset ↗Kaggle · emmarex/plantdiseaselines:314-336Plant phenotyping relevance matchCrossref · checked 5 Sept 2026
Introduction Accurate estimation of aboveground biomass (AGB) is essential for monitoring pasture productivity and supporting sustainable management of integrated crop–livestock (ICL) systems. We hypothesized that integrating multispectral, thermal, and canopy-structural information derived from unmanned aerial vehicles (UAVs) would improve AGB prediction relative to spectral information alone, and that Generalized Additive Models for Location, Scale and Shape (GAMLSS) would accommodate seasonal heteroscedasticity while maintaining predictive performance comparable to Random Forest (RF) and Support Vector Machine (SVM) models. Methods We collected 280 destructive biomass samples from two ICL paddocks and one continuously grazed pasture in the Brazilian Cerrado between 2022 and 2024. Twenty-four UAV-derived predictors, including spectral bands, vegetation indices, canopy surface temperature, and canopy height, were evaluated using repeated five-fold cross-validation. Model transferability was assessed by withholding one management paddock at a time. Results and discussion Under repeated five-fold cross-validation, GAMLSS achieved the lowest prediction error (R² = 0.69 ± 0.01; RMSE = 2.15 ± 0.04 Mg ha⁻¹), followed closely by SVM (R² = 0.68 ± 0.01; RMSE = 2.19 ± 0.03 Mg ha -1 ); RF showed lower accuracy (R 2 = 0.53 ± 0.01; RMSE = 2.63 ± 0.02 Mg ha -1 ). In the paddock-transferability assessment, GAMLSS also showed the lowest error (R 2 = 0.63 ± 0.04; RMSE = 2.34 ± 0.26 Mg ha -1 ). For GAMLSS, the complete multisensor configuration reduced RMSE by 6.2% compared with the spectral-only configuration. The selected model was used to generate spatially explicit maps of AGB and standing aboveground biomass carbon, estimated from the mean measured carbon concentration of forage biomass. Integrating multispectral, thermal, and structural UAV data with distributional regression improves AGB estimation and enables spatial monitoring of tropical pastures under contrasting management conditions.
Biomolecular condensates that persist through cell division must be reorganized and inherited, yet it remains unclear whether subtle defects before division are associated with later organelle or growth phenotypes. We examined the Chlamydomonas reinhardtii pyrenoid, a liquid-like condensate that concentrates ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco), the photosynthetic CO2-fixing enzyme. As part of the algal CO2-concentrating mechanism, the pyrenoid raises CO2 availability around Rubisco. We generated an RBCS1-mGold Rubisco reporter and developed an unsupervised image-analysis pipeline combining a convolutional autoencoder and a one-class support vector machine. Using 4,905 wild-type single-cell images, augmented 22-fold to 107,910 image instances, we defined the range of normal pyrenoid morphology. A combined machine-learning and visual screen of approximately 21,000 insertional mutants yielded 17 pyrenoid integrity mutants (pim1-pim17). Differential reconstruction-error maps highlighted local deviations from the wild-type reference, including phenotypes difficult to classify by eye. Four-dimensional live imaging showed defects in matrix dispersal, partitioning of Rubisco-containing foci, or pyrenoid recondensation in multiple pim strains. Growth assays identified broad defects and phenotypes that became more apparent as CO2 supply decreased. Insertion-site mapping nominated candidate loci, including STT7, which encodes a chloroplast kinase best known for regulating photosynthetic light harvesting. Independent STT7-edited lines lacked detectable STT7 accumulation and showed pyrenoid-region reconstruction-error patterns, supporting an association between impaired STT7 function and altered pyrenoid morphology. These findings show that unsupervised image screening can extend forward genetics to subtle pyrenoid phenotypes accompanied by mitotic remodeling or growth defects.
Reproduction assets foundThe paper's custom machine-learning analysis scripts (CAE–OC-SVM pyrenoid screening pipeline) are explicitly stated to be publicly available on the authors' GitHub repository. Other data (microscopy files, anomaly scores) are only available upon request, so they do not qualify as public assets.Code · publicCustom scripts used for the machine-learning analyses are publicly available at https://github.com/Yamano-Lab/2025_Machine_Learning-based_screening .Open asset ↗Yamano-Lab/2025_Machine_Learning-based_screeninglines:103-119Plant phenotyping relevance matchCrossref · checked 15 Sept 2026
Ilya A. Grishin · Gennady I. Afanasyev · Valeri I. Terekhov
Field / plotLiDAR / point cloudStem / branchWhole plant / canopy / plot / fieldClassification2D/3D reconstructionSegmentationArchitecture / morphology / geometry
The separation of a forest plot into individual trees and the automatic extraction of their stem structures from terrestrial laser scanning data are complicated by dense stands, occlusions, and the diversity of biomorphological forms. Existing algorithms usually solve scene partitioning, voxel classification, and tree growing as independent tasks, which leads to error accumulation at subsequent processing stages. This paper proposes a unified model for spatial fragmentation and biomorphological forest segmentation comprising three interrelated stages: scene partitioning by estimated stem coordinates using a Voronoi diagram, probabilistic voxel- or point-level classification, and bottom-up tree growing guided by spatial connectivity and stem membership criteria. For the semantic module, tabular, volumetric, and point-based approaches are compared: gradient boosting with layer-by-layer inference, TabNet, a three-dimensional convolutional neural network, PointNet2, and two-stage pipelines in which gradient boosting builds an initial stem mask for subsequent neural segmentation. The experiment was conducted on 567 mixed-species trees. Considering both quality and computational performance, the {CatBoost; CNN3D} pipeline was selected as the preferred solution, achieving AUC = 0.9966 and IoU = 0.9831. The obtained results show that combining interpretable layer-by-layer classification with subsequent spatial analysis improves the quality of stem structure reconstruction, which is important for automatic forest inventory tasks.
Tomato leaf diseases must be identified early and accurately in order to reduce output loss and advance sustainable agriculture. Deep learning models have shown encouraging results in the identification of plant diseases, but their high processing requirements and inability to adjust to field-specific limitations sometimes make it difficult to implement them in real-world applications. For the purpose of accurately and efficiently classifying tomato leaf diseases, we present a convolutional neural network based on AlexNet that is lightweight and field-aware. Our proposed model is designed with less complexity than traditional architectures and is able to perform inference faster without sacrificing accuracy, making it suitable for real-time implementation in low resource agricultural environments. The algorithm was trained and tested on a dataset of 7704 augmented images of tomato leaves from 7 different disease categories. The Lightweight AlexNet achieved accuracy comparable to VGG variants and outperformed deeper models such as ResNet (96.81%) with lower parameter overhead. It was trained from scratch using TensorFlow & Keras on 100×100 pixel inputs, achieving a training accuracy of 99.15% and a validation accuracy of 99.74%. Moreover, an effective structure of the model enables the installation on edge devices, which offers a scalable precision farming solution. Our work helps to bridge the gap between deep learning research and real-world application in agriculture, allowing the development of real-field, resource-efficient disease detection systems.
Abstract Living tissues contain dynamic biochemical information that is difficult to capture with conventional hyperspectral microscopes because sequential spectral acquisition is poorly matched to in vivo molecular processes that evolve during measurement. Here we introduce a task-specific optical encoding framework for video-rate molecular inference in living plant tissue. The system integrates a passive spectral encoder, implemented here as a low-angle scattering LDPE layer, into a 22-mm miniaturized probe and learns a supervised mapping from ultraviolet-excited autofluorescence measurements to biomolecular abundance maps. Unlike conventional pipelines that first reconstruct hyperspectral datacubes and then perform spectral unmixing, the deployed system directly estimates endogenous molecular contrast associated primarily with lignin and chlorophyll in poplar tissue, with additional suberin-associated contrast evaluated in suberin-rich tissue. This reframing makes the measurement task biomolecular inference rather than spectral reconstruction, enabling biochemical mapping under low-photon autofluorescence conditions while reducing data burden and computational latency. In living poplar stems, the platform captures autofluorescence-derived videos of embolism propagation and wound-induced biochemical remodeling, dynamic processes for which sequential spectral acquisition can introduce temporal mixing because the molecular contrast evolves during the scan itself. The system also resolves genotype-dependent reductions in lignin-associated autofluorescence in engineered poplar lines. Direct molecular inference improves biomolecular estimation relative to a reconstruction-based pipeline, while probabilistic decoding provides uncertainty estimates. These results show that compact passive spectral encoding, when optimized for biological inference rather than datacube recovery, enables deployable, label-free molecular videography of living plant tissue dynamics after task-specific calibration.
Plant diseases can reduce crop quality and productivity, making early detection an important aspect of modern agriculture. Recent advances in deep learning, particularly Convolutional Neural Networks (CNN), have shown promising performance in image-based plant disease classification. This study proposes an explainable deep learning approach for multi-class plant disease classification using ResNet50 and EfficientNetB0 combined with Grad-CAM visualization. The experiments were conducted using the PlantVillage dataset consisting of 15 classes of healthy and diseased plant leaves.The research process included image preprocessing, data augmentation, transfer learning, model training, performance evaluation, and explainability analysis. The dataset was divided into training and validation sets with a ratio of 80:20. Model performance was evaluated using accuracy, loss, confusion matrix, precision, recall, and f1-score metrics. Experimental results showed that ResNet50 achieved the best performance with an accuracy of 92% and a validation loss of 0.19, outperforming EfficientNetB0 which obtained 76% accuracy and 0.82 validation loss. The classification report demonstrated that ResNet50 provided more stable and consistent predictions across most disease classes. Furthermore, Grad-CAM visualization successfully highlighted disease-relevant regions such as lesions, discoloration, and damaged leaf areas, improving the interpretability of the CNN model. The findings indicate that the combination of ResNet50 and Grad-CAM is effective for plant disease classification and provides better explainability for deep learning-based agricultural applications.
To establish an accurate and interpretable prediction framework for soybean lodging grade and clarify the core regulatory traits and differentiated driving mechanisms of soybean lodging under high-density drip irrigation cultivation, 356 spring soybean germplasm accessions were used as experimental materials in this study. Morphological and mechanical traits including plant height (PH), stem pulling force (SPF), internode number (IN) and petiole length (PL) were measured over two consecutive years of field phenotyping. Two composite evaluation indices, plant height/stem pulling force ratio (PH/SPF) and plant height/internode number ratio (PH/IN), were further constructed. Four machine learning algorithms were adopted to develop multi-classification models for soybean lodging grade prediction. SHAP analysis combined with three global sensitivity approaches (perturbation analysis, Sobol’ method and Morris screening) was applied to decipher the regulatory patterns of key traits. The results showed that lodging grade significantly affected soybean grain yield and explained 25–28% of the phenotypic yield variation; yield reduction tended to plateau under severe lodging. Compared with single indicators such as SPF and PL, the two derived composite indices could stably distinguish soybean accessions with different lodging grades and exhibited stronger discriminatory power. Model comparison revealed that the XGBoost model achieved optimal prediction accuracy and generalization stability for lodging grade, with a weighted F1-score of 95.34% on the test set, significantly outperforming the conventional linear model. Interpretability analysis demonstrated that the PH/IN, PH, and PH/SPF acted as the primary positive traits promoting lodging, while SPF was the sole protective trait. Driving factors of lodging presented obvious gradient heterogeneity: mild lodging was dominated by the imbalance of plant architecture ratio, whereas severe lodging was governed by the cumulative effects of PH and IN. Strong interactions existed among all measured traits. The interpretable machine learning framework established in this study can provide theoretical support and technical references for lodging-resistant germplasm screening and targeted plant architecture regulation for densely planted soybean under drip irrigation systems.
Herbarium specimens are physical, verifiable records that form the basis of taxonomic knowledge and biodiversity research. Their large-scale digitization has produced extensive collections of high-resolution images and associated specimen metadata, creating conditions in which artificial intelligence (AI) can play an important role in plant taxonomy, collection management, and ecological research. Early AI applications have primarily focused on automated species identification based on individual specimen images. Although increasingly accurate, such approaches remain limited by their emphasis on single-specimen label prediction and by treating identification outputs as final analytical decisions. Recent methodological advances-including segmentation-based preprocessing, automated trait extraction, structured extraction of label data, detection of potentially misidentified specimens, and multimodal integration of visual, textual, and genetic information-extend AI applications beyond species identification toward broader analytical frameworks, encompassing taxonomic interpretation as well as ecological and biodiversity research. In these approaches, specimens are placed within a shared analytical space, and identification results are used to support comparisons across multiple specimens rather than being treated as final decisions for single individuals. This multi-specimen perspective enables quantitative examination of species boundaries, morphological variation, data inconsistencies, and taxonomic stability within curated collections. In this context, AI serves not as an ultimate decision-maker but as a decision-support tool embedded in expert-guided workflows and biodiversity knowledge infrastructures. These developments can be summarized as Integrative Taxonomic AI, an approach that employs learned morphospaces to interpret and refine taxonomic categories by integrating multimodal evidence and curated specimen data under expert guidance.
Microplastics (MPs) are persistent and ubiquitous contaminants in aquatic ecosystems, yet their interactions with submerged aquatic plants remain poorly understood. While MP-induced phytotoxicity has been extensively investigated in terrestrial plants, quantitative evidence for MP uptake and internal accumulation in submerged species is still limited. In this study, we investigated the phytotoxicity and accumulation patterns of fluorescent microplastics (FMPs) in two submerged aquatic plants, Bacopa lanigera and Rotala indica, using fluorescence spectroscopy. Plants were exposed to FMPs of two particle sizes (50 nm and 1 µm) across three exposure concentrations (0.001%, 0.01%, and 0.05%). Plant growth, chlorophyll content, fluorescence emission, and FMP accumulation were systematically evaluated. Our results demonstrated clear size- and concentration-dependent responses. Smaller particles (50 nm) showed significantly higher uptake and induced stronger phytotoxic effects than 1 µm particles, with pronounced growth inhibition and chlorophyll reduction observed at the highest concentration (0.05%). Fluorescence-based analysis enabled quantitative estimation of both surface-associated and internalized FMPs within plant tissues. Maximum surface accumulation reached 207 ppm, while internal (cross-sectional) accumulation reached up to 75 ppm, regardless of plant species. Under the respective experimental conditions, B. lanigera exhibited higher estimated FMP accumulation, whereas R. indica showed greater growth inhibition. These findings provide quantitative evidence of microplastic uptake and internal accumulation in submerged aquatic plants and highlight particle size as a critical determinant of phytotoxicity. Moreover, this study establishes a fluorescence-based methodological framework for estimating microplastic concentrations in aquatic plant tissues, contributing to improved ecological risk assessment of microplastics in freshwater ecosystems.
Abstract Purpose : The system for diagnosing diseases in Solanaceae crops (SolanAPP), including tomatoes, potatoes, peppers, and eggplants, represents a promising tool for supporting decision-making in agricultural fields using AI. This system reconciles the computational intensity of multitasking models with the infrastructural limitations of rural environments, thereby increasing digital literacy. Its architecture is based on two fundamental pillars: (i) autonomous, offline operation for the detection and classification of diseases in Solanaceae crops; (ii) a georeferenced epidemiological surveillance network with agricultural recommendations for crop monitoring. Methods : The core diagnostic process combines crop-specific semantic segmentation and disease classification models exported to TensorFlow Lite, enabling on-device visual inference and pixel-level severity estimation. An optional online layer integrates Groq’s large language model (LLM)-based reasoning and Firebase services to generate structured agronomic explanations and facilitate the creation of georeferenced community reports when a connection is available. Preprocessing and management of the dataset were performed using the Roboflow platform. The mobile app was developed natively in Kotlin. Model performance was rigorously evaluated using accuracy, recall, F1 score, mean IoU, mPA, inference latency, model size, and decision matrix. The optimal model for the crop was selected using Simple Additive Weighting (SAW). Finally, the overall framework quality and usability were evaluated in Cuba through a user validation survey using a 5-point Likert scale and aligned with the ISO/IEC 25010 software quality model. Results : The model that yielded the best results for most crops was DeepLabV3+ with MobileNetV2, which achieved a classification accuracy of over 97\% while operating with lower inference latency. Beyond individual diagnoses, the system incorporates a collaborative georeferencing feature that allows users to share observations and precise geographic coordinates of detected pathologies to facilitate regional epidemiological monitoring. The user satisfaction survey yielded a satisfaction rating of 4.5/5, with users highlighting the importance of offline diagnosis. Conclusion : Plant disease diagnosis using computer vision can support earlier intervention in resource-constrained agricultural settings, but practical deployment requires models that are accurate, lightweight, interpretable, and usable under limited connectivity. SolanAPP, an offline-first Android framework for detecting foliar pathologies in Solanaceae crops, not only establishes a framework for disease identification in complex natural environments but also provides a theoretical and practical foundation for automated agronomic treatment recommendations and community-based crop surveillance. Impact SolanAPP is a free framework that supports the synergy between multitask deep learning for offline disease diagnosis and LLM-driven reasoning for decision-making in the field. Beyond the quantitative metrics obtained from the selected models, the deployment of SolanAPP in rural contexts serves a fundamental socio-technical purpose: it acts as a catalyst for open access, digital literacy, and agronomic decision-making under unfavorable development conditions. It also represents a strong effort to foster a collaborative epidemiological surveillance network in the agricultural sector. Although it faces challenges, such as the use of field images for model training, this framework marks a promising step in the deployment of edge AI, balancing technical accuracy with practical utility.
Abstract Plant diseases represent major constraints on agricultural productivity, often resulting in significant yield losses. This study presents FitoView, a cloud-based mobile decision support system that integrates deep learning with real-time weather forecasting for sustainable plant disease management. Demonstrated through a case study on Cercospora leaf spot in chili pepper, the system employs custom YOLOv8 models for automated disease detection, classification, and pixel-level severity quantification, combined with meteorological data from the OpenMeteo API. The core innovation lies in an integrated decision matrix that considers three dimensions: AI-assessed disease severity, 48-hour climatic risk forecasts, and optimal spraying conditions, generating four contextualized management scenarios with tailored re-evaluation periods (3-10 days). OpenMeteo API validation across four cities in Sergipe demonstrated very strong correlations, with Pearson coefficients (r) of 0.90-0.97 for temperature, 0.81-0.95 for humidity, and 0.92-0.95 for solar radiation, corresponding to R² values of 0.65-0.94. The YOLOv8 object detection model achieved perfect precision (100%) and macro-averaged recall of 89% across all disease classes, with Cercospora leaf spot detection reaching perfect metrics (100% precision, recall, and F1-score). Field validation in Lagarto, Sergipe, confirmed the system’s practical use: it accurately detected Cercospora leaf spot, estimated severity, and, combined with climatic risk, generated recommendations for alternative treatment and short-term re-evaluation. The Progressive Web Application architecture, deployed on a Cloud Platform, ensures accessibility without installation requirements, while the modular design enables scalability to additional crops and diseases, representing a significant advancement toward democratizing AI-powered precision agriculture tools for smallholder farmers in Brazil.
Maize ( Zea mays L.) is a highly adaptable crop grown worldwide across diverse climates and management practices, with uses across multiple sectors. Consequently, the traits prioritized in maize research and breeding programs vary depending on the specific objectives. Core traits, however, such as flowering time, plant and ear height, and stalk and root lodging, which are important for evaluating and improving the performance and stability of maize genotypes, are routinely evaluated across breeding programs, regardless of their goals. Standardized measurement of these core traits is essential to ensure data reliability and comparability, enabling the integration of phenotypic data across different experiments. Such efforts ultimately support better decision-making and accelerate the development of improved maize genotypes. This is particularly important in public sector programs, where large-scale evaluations, critical for assessing the value of specific genotypes, are often only feasible through collaboration across programs. Here, we provide a protocol for the standardized collection of phenotypic data, specifically focusing on how to measure core traits in maize field trials. These methods promote consistency and accuracy in the evaluation of these traits, and support communication and coordination among groups in the public sector and other research settings. Further, such standardization facilitates the integration and comparison of data across programs, enabling robust longitudinal and multienvironment analyses.
Florian Tanner · Judith Atieno · Sara N. Blake · M. Krysinska-Kaczmarek · Chris Brien · Mohsen Khani · K.R. Clarke · Darren Plett · Jennifer Davidson · Bettina Berger
Abstract Ascochyta blight is a widely occurring chickpea fungal disease that can cause severe yield loss. Breeding for crop resistance benefits from high‐throughput evaluation of plant–pathogen interactions in genotypes which can serve as sources of resistance. Current practice for the evaluation is human visual scoring of disease symptoms, which is limited in throughput and precision. Here, we developed open‐source sensor‐based phenotyping methods using red, green, blue (RGB) and multispectral imaging to measure resistance components and predict disease severity classes in chickpea and wild relatives grown outdoors over three seasons. Pots were imaged at multiple time points with a ground‐based platform, providing 86,792 RGB and 8199 multispectral images. Lesion count was estimated with YOLOv5 (You Only Look Once version 5) object detection (F1 score = 0.27–0.30), fractional green canopy cover was estimated from RGB images, and vegetation indices were extracted from multispectral images. A model trained on growth rates of fractional green canopy cover normalized to control genotypes could predict disease severity classes with an accuracy of 65% –81 % ( 0.43–0.59) on unseen data from three different seasons. The developed methods provide a pathway to predict visual disease severity scores and support the breeding of crops for disease resistance. They may also be used to characterize disease progression, to find underlying resistance mechanisms, and for early disease detection.
Abstract This study evaluated the potential of UAV-derived RGB spectral indices to predict fiber quality traits in cotton genotypes. Nineteen genotypes were assessed under a randomized complete block design, and RGB imagery acquired at full flowering was used to calculate GLI, NGRDI, SCId, and SI. Significant genetic variability and moderate-to-high heritability were observed for both fiber traits and spectral indices. GLI was positively associated with the Spinning Consistency Index, whereas NGRDI was associated with fiber length uniformity. Regression models showed moderate predictive ability (R² LOOCV between 32.4–35.4%; and accuracy between 0.57–0.60). GLI and NGRDI demonstrated potential as complementary tools for large-scale phenotyping and preliminary genotype selection, although they do not replace conventional fiber quality analyses. Further studies across additional developmental stages are needed to improve prediction accuracy.
The rapid detection and continuous monitoring of crop health and disease outbreaks are critical components of modern precision agriculture, essential for maintaining global food security. Traditional field-based scouting methods, while accurate, are often labor-intensive, time-consuming, and limited by spatial coverage, making them inadequate for large-scale agricultural operations. Remote sensing (RS) technologies—spanning satellite imagery, drone-based aerial platforms, and proximal sensors—offer a powerful, non-destructive, and scalable alternative for capturing high-resolution spectral and temporal data. This paper provides a comprehensive evaluation of current remote sensing applications in crop health monitoring and disease surveillance. We analyze how vegetation indices derived from multispectral and hyperspectral data, such as NDVI and red-edge parameters, serve as sensitive indicators of physiological stress and pathogen infection, often manifesting before visible symptoms appear. Furthermore, we explore the integration of machine learning and artificial intelligence algorithms in automating disease identification and severity mapping. By synthesizing recent advancements in sensor technology and data analytics, this paper demonstrates that remote sensing is indispensable for proactive, site-specific management. The findings emphasize that a multi-scale RS approach—integrating broad-scale satellite monitoring with high-resolution drone sorties—enables farmers to optimize input efficiency, minimize yield losses, and enhance the overall resilience of agro-ecosystems against biotic and abiotic stressors.
Plant disease and plant stress early warning systems have advanced through deep learning, remote sensing, digital phenotyping, disease forecasting, and sensor networks. Detection accuracy, precision, recall, F1-score, and area under the curve remain indispensable, but they are insufficient for judging whether a warning can support timely and proportionate phytoprotection under field variability. This Mini Review argues that intelligent plant health warning systems should be evaluated not only as prediction models, but also as safety-relevant decision-support systems embedded in biological, agronomic, and operational contexts. We first relate AI-based detection to established plant disease forecasting and decision-support traditions, including weather-based models, epidemiological forecasting, and integrated disease management. We then adapt selected safety-assurance concepts, including risk assessment, failure mode and effects analysis, Bow-tie reasoning, warning-threshold governance, reliability analysis, resilience thinking, and response closure, to host-pathogen-environment warning chains. The proposed framework links AI or sensor outputs with pathogen biology, host susceptibility, environmental conduciveness, inoculum pressure, uncertainty assessment, risk classification, threshold decisions, human or automated verification, intervention, and feedback learning. Illustrative crop-pathogen scenarios, including wheat rust, rice blast, potato late blight, grapevine downy mildew, and citrus greening, show how safety assurance can complement existing forecasting and decision-support systems rather than replace them. The framework remains conceptual, and whether these added assurance functions improve existing warning systems requires comparative evaluation under field conditions. Future systems should be evaluated through detection performance and response-oriented indicators such as lead time, calibration, false-alert burden, missed-warning rate, response completion, disease suppression, economic value, and learning after field action.
Precision agriculture increasingly requires intelligent systems capable of integrating multimodal sensing with transparent decision support to enable timely and reliable crop management. This study proposes a hybrid intelligent IoT framework integrating environmental monitoring, wearable plant physiological sensing, AI-based pest-monitoring, machine-learning-based prediction of crop physiological stress, and explainable fuzzy rule-based decision support into a unified architecture for crop stress assessment. A novel Physiological Stress Index (PSI) was developed by combining vapor pressure deficit, relative humidity, Delta-T, leaf capacitance, and relative irradiance to provide an interpretable indicator of crop physiological stress. The proposed framework was experimentally validated under real field conditions using a commercial environmental monitoring station, wearable leaf sensors, AI-enabled pest-monitoring devices, and cloud-based analytics. Correlation analysis confirmed strong relationships between PSI and the principal environmental variables (VPD: r = 0.980, Delta-T: r = 0.990, RH: r = −0.961), demonstrating the internal consistency and sensitivity of the proposed index. At the 15 min forecasting horizon, Linear Regression and Gradient Boosting demonstrated virtually identical performance: Gradient Boosting achieved a marginally lower RMSE and higher R2 (RMSE = 0.0273; R2 = 0.9810), whereas Linear Regression achieved a slightly lower MAE (MAE = 0.0186). At the 1 h forecasting horizon, Gradient Boosting achieved the strongest performance (R2 = 0.9034), indicating increasing relevance of nonlinear modelling at longer prediction horizons. The proposed framework demonstrates the feasibility of combining multimodal sensing, machine learning, explainable artificial intelligence, and edge-enabled IoT technologies to support proactive, transparent, and intelligent precision agriculture.
Abstract Disease management aims to protect crop yield and quality and reduce economic losses caused by plant pathogens. Consequently, reducing disease is a central objective of applied plant pathology. However, what constitutes effective disease control, and how it is measured and analyzed, varies substantially among studies. We conducted a systematic methodological review to characterize how plant disease control has been evaluated in the plant pathology literature over the past 15 years. We searched selected plant pathology journals for articles containing "control" in their titles and used an artificial intelligence-assisted workflow, followed by human verification, to extract and standardize information on experimental settings, disease measurements, measurement scales, and statistical analyses. The final dataset comprised 340 articles representing diverse host-pathogen systems and experimental environments. Disease control was evaluated using a wide range of response variables, most commonly disease severity and incidence, with substantial heterogeneity in measurement scales and sampling practices. Despite this diversity, statistical analysis was remarkably uniform: 79.4% of articles relied exclusively on ANOVA-based approaches. Among studies using ordinal disease scales, 73.3% included ANOVA in the analysis, whereas only 10.6% explicitly reported data transformation. Mean-separation procedures were also common, particularly Tukey, Fisher's LSD, and Duncan's multiple range test; Duncan's test was reported in 20.3% of all articles and varied markedly among journals. Our findings reveal a marked contrast between diversity in how plant disease control is measured and the narrower range of methods used to analyze those measurements. Greater alignment among biological meaning, measurement properties, experimental design, and statistical analysis could improve transparency, comparability, and interpretation in disease-management research.
Abstract. In the face of escalating global food demand and increasing climate variability, precise and granular crop yield monitoring is indispensable for maintaining regional agricultural stability. However, current deep learning approaches for yield estimation are severely constrained by their heavy reliance on massive in situ labeled data, which limits their application in data-scarce regions. Furthermore, these models often overlook the essential temporal evolution logic of yield formation and lack a systematic discussion regarding the contribution patterns of different feature dimensions, resulting in a black-box nature of the underlying model mechanisms. To address these challenges, this study proposes a field-label-free training framework for maize yield estimation that couples mechanistic model with deep learning. The framework's core strength lies in a physiologically complete simulation database, using the WOFOST model to exhaustively cover 30 years of climate variability and habitat combinations across Northeast China (1.24 × 106 km2). A Gated Recurrent Unit (GRU) network was then introduced for end-to-end modeling, accurately capturing the energy accumulation trajectory from vegetative to reproductive growth. Validation against 458 independent ground points (2022–2024) demonstrated robust generalization with an R2 of 0.69, an RMSE of 1.21 t ha−1, and an RRMSE of 13.73 %, despite using no ground data for training. Our analysis revealed that integrating photosynthetic intensity (LAImean), duration (LAD) and peak features (LAImax) across growth stages is critical for accuracy, while omitting early-stage features significantly impairs the model's ability to capture cumulative growth effects. Furthermore, the model successfully captured the spatiotemporal yield anomalies caused by the 2023 typhoon and flooding events. Ultimately, this study generated a 10 m resolution maize yield dataset (2019–2024) for Northeast China. The dataset exhibits consistent interannual stability, with the RRMSE ranging from 7.98 % to 12.92 % and the R2 remaining above 0.44 at the city level. By deeply coupling mechanistic simulation with data mining, this dataset provides detailed support for optimizing agricultural production and guiding farming practices. The Northeast China Maize Yield 10 m dataset is openly available at https://doi.org/10.5281/zenodo.19547014 (Hu et al., 2026).
Reproduction assets foundThe paper's core output, the NortheastChinaMaizeYield10m maize yield dataset (2019–2024) with accompanying uncertainty layers, is openly deposited on Zenodo with an explicit availability statement and DOI. No author analysis code or trained model checkpoints are stated as publicly available.Dataset · publicThe Northeast China Maize Yield 10 m dataset is openly available at https://doi.org/10.5281/zenodo.19547014 (Hu et al., 2026).Open asset ↗Zenodo · 10.5281/zenodo.19547014lines:158-191Plant phenotyping relevance matchEurope PMC · checked 14 Sept 2026
Northeast China's Geng rice (Oryza sativa subsp. japonica) dominates the high-value rice markets in China due to its superior eating quality. However, current evaluation methods rely on either labor-intensive, subjective sensory protocols or low-accuracy, calibration-heavy near-infrared spectroscopy (NIRS), constraining breeding for high eating quality and market development. Here, we report a vision-based deep learning framework combining multi-population fine-tuning with industrial vision-language model (VLM) pre-training for Geng rice eating quality prediction. Trained on natural and recombinant inbred (RI) population datasets, our optimal model (Model 4) showed high cross-population stability. It achieved R 2 values of 0.98, 0.57, and 0.61 in a natural population validation set (35 cultivars), an independent DA-RI population (201 lines), and a randomly collected set (30 Northeast and 28 Southern cultivars), respectively, consistently outperforming the widely used Satake STA1B analyzer. Furthermore, our approach enabled the mapping of a novel, robust quantitative trait locus, qIVOE7, for Geng rice eating quality on Chromosome 7. Further analysis suggested that Model 4 appears to rely on the Hue dimension of the HSV color space for its predictions. This framework provides a high-accuracy prediction model and an easy-to-use tool for rice eating quality evaluation, accelerating high-quality rice breeding as well as the development of the high-quality rice market.
Plant diseases destroy 20–40% of global food production annually, posing a critical threat to food security for a projected population of 9.7 billion by 2050. Conventional diagnostic approaches relying on expert visual assessment are slow, costly, and unsuitable for modern agricultural scales. While deep convolutional neural networks demonstrated early promise, single-modality, image-centric systems consistently fail under real-world field conditions characterized by variable lighting, co-occurring infections, and cultivar diversity. This review synthesizes a decade of progress across four interconnected frontiers: the evolution of deep learning architectures for plant disease detection; the adaptation of foundation models including CLIP, SAM, and DINOv2 to agricultural contexts; the development of multimodal fusion frameworks integrating imagery, environmental, genomic, and hyperspectral data; and the transition from static disease diagnosis to descriptive comparison of reported metrics, which suggested that multimodal approaches frequently reported improved diagnostic performance relative to corresponding single-modality baselines, although direct cross-study comparison was limited by methodological heterogeneity. A systematic review following PRISMA guidelines identifies eligible comparative studies. Descriptive comparison of reported performance metrics across these studies indicated that multimodal approaches generally achieved higher accuracy and sensitivity than single-modality models, particularly for pre-symptomatic disease detection. Eight critical research gaps are identified, including the absence of a unified agricultural foundation model and limited climate-aware forecasting under non-stationary climate projections. A structured research agenda is proposed to accelerate translation from laboratory performance to globally equitable, field-deployable crop protection systems.
Abstract In modern agriculture, it is essential to identify the early symptoms of plant diseases and to accurately maintain the productivity of the crop and reduce economic losses. Foliar diseases are a special concern in peach because they can cause yield as well as quality if not timely detected. Artificial intelligence, machine learning and deep learning are some of the advanced technologies that are gaining great importance in today's agriculture, especially with the image analysis applications. In this study, a deep learning method for automated detection and classification of three peach leaf diseases and healthy class was presented based on image data. The dataset were taken at different phenological and disease stages under temperate conditions in Kashmir with four classes Healthy, Leaf Curl, Shot hole and Rust. Three convolutional neural networks (CNNs) architectures were applied, VGG-16, ResNet 50 and Xception were trained using transfer learning and Inception-V4 was trained from scratch for a comparative study of the learning strategies. Data augmentation techniques were applied to improve generalization. Results show that all models were able to learn disease specific features well. The result of Inception-V4 was found to be highest with 97.50%, followed by ResNet-50 with 94.49%, VGG-16 with 92.04% and Xception with 75.95%. The results of transfer learning-based architectures were also good and competitive but the best results obtained from the Inception-V4 architecture reveal its capability in modelling complex visual patterns. The results highlight the potential of deep learning techniques for early detection of diseases in peach, supporting precision agriculture and better disease management.
Accurate and non-destructive assessment of drought stress is important for improving lettuce production and supporting timely crop management. This study presents a detection-guided deep learning framework for plant-level drought-stress assessment in hydroponically grown lettuce using bird’s-eye-view RGB images. The study further investigates whether canopy segmentation can improve classification performance by reducing irrelevant background information. The framework was evaluated using 2190 images collected across three independent cultivation cycles in which drought stress was induced by isolating the plant root zones from the nutrient solution. In the first stage, YOLO-based object detection was used to localize individual plants, with YOLO26m achieving the highest detection performance of 99.4% mAP@0.5. The detected regions were subsequently used as spatial prompts for zero-shot canopy segmentation using the Segment Anything Model (SAM), with SAM ViT-B achieving a mean IoU of 0.9864. Six convolutional, transformer-based, and hybrid classification architectures were then evaluated independently using YOLO-cropped and SAM-segmented plant images. Segmented inputs consistently improved classification performance, with MaxViT-S achieving the highest binary test accuracy of 96.3%. The framework further distinguished time-defined pre-stress, early-stress, and late-stress periods with an accuracy of 92.4%. Plant-level generalization was further assessed using six-fold leave-one-plant-out cross-validation, resulting in a mean test accuracy of 90.25 ± 1.78% on unseen plants. These findings demonstrate that RGB-based plant-level analysis can support non-destructive drought-stress assessment and that canopy segmentation improves classification by reducing background influence.
Abstract A systematic review based on 174 Scopus-records of studies using YOLO-type one-stage detectors for detecting citrus fruits, their count, and yield estimation followed PRISMA guidelines 2020. The title/abstract-screening process, done in duplicate (κ=0.920) yielded 90 included study-records, followed by two further post-hoc exclusions. Each study in the 80 reporting on fruit-level detection showed an average precision of 89.6%, recall of 85.9%, and mAP@0.5 of 91.0%. However, coverage for any individual metric rarely exceeded half of the studies, and only 13% were able to report the more stringent mAP@0.5:0.95. Both YOLOv8 and YOLOv5 were each utilized as the backbone architecture by approximately 22.2% of the studies. From 2025, YOLOv11 has also been emerging. Half of all studies modified architectural components including attention modules, lightweight architectures, and variants of IoU loss functions. Original contributions are generally concentrated in downstream tracking, sensor fusion, and yield modeling rather than the detector itself. A custom-made seven-domain risk of bias tool was developed and utilized by two reviewers who arbitrated discrepancies (91.5%). Results showed that all but one of the reviewed studies had a high level of risk due to almost universal lack of statistical validation and limited dataset diversity; a sensitivity analysis excluding the most risky studies left the performance profiles nearly identical. We conclude that the field has converged around a common technical toolkit but continues to lack standardized benchmarks, multispectral data, and rigorous field-deployment validation.
Potato, one of the world's most important staple food crops, is highly susceptible to various foliar diseases that significantly affect its productivity and pose a serious threat to food security, thereby contributing to economic losses and impacting farmers’ income. Therefore, early and accurate detection is essential. Conventional detection methods rely primarily on manual observation, which is time-consuming and requires specialized personnel. In this study, five convolutional neural network (CNN) architectures were evaluated for the automatic classification of potato leaf diseases, including pretrained models (ResNet50, MobileNet, and VGG16) and models trained from scratch (AlexNet and LeNet-5). The dataset was constructed by integrating and selecting images from publicly available Kaggle repositories, resulting in a total of 6,691 images distributed across five classes: early blight, late blight, potato leafroll virus (PLRV), mosaic virus (PVY), and healthy leaves. Multiple experiments were conducted by varying hyperparameters such as batch size, optimizers, and the number of training epochs. The results show that VGG16 achieved the best performance, with an accuracy of 99.87%, outperforming the other architectures. Additionally, a mobile application based on the optimal model was developed for real-time detection. These findings demonstrate the potential of deep learning for intelligent and scalable agricultural diagnostic systems.
Background Global soybean production is constrained by scarce arable land, and standardized evaluation tools remain lacking for natural mixed saline-alkali stress, the predominant abiotic stress under field conditions. Objective This study aimed to establish a comprehensive saline-alkali tolerance evaluation system for soybean germplasms via integrated multivariate statistical methods, and screen core and auxiliary indicators for efficient germplasm identification. Methods Seventy-one soybean germplasms were tested under 90 mmol/L mixed saline-alkali stress (NaCl:Na 2 SO 4 :NaHCO 3 :Na 2 CO 3 = 1:9:9:1, pH 8.2) simulating natural saline-alkali soil. We quantified the saline-alkali tolerance coefficients (SATC) of 13 morphological and physiological indicators, followed by coefficient of variation (CV), principal component analysis (PCA), subordinate function, cluster analysis and regression modeling. Results Significant inter-germplasm variations in saline-alkali tolerance were detected, and indicators with CV > 0.35 ( e.g ., root length (RL), root fresh weight (RFW)) were screened as primary indices. PCA extracted five principal components with 87.18% cumulative variance contribution, and the integrated analytical pipeline categorized germplasms into five tolerance grades: eight highly tolerant, 24 moderately tolerant, 13 generally tolerant, 16 sensitive and 10 highly sensitive accessions. A high-precision prediction model was constructed ( D = 0.290 X 1 - 0.026 X 2 + 0.438 X 3 + 0.402 X 4 + 0.180 X 5 + 0.153 X 6 + 0.813 X 7 - 1.123; R 2 = 0.998, where X 1 - X 7 represent the SATC of germination rate (GR), RL, RFW, total fresh weight (TFW), shoot dry weight (SDW), root dry weight (RDW), and total dry weight (TDW), respectively). A Chi-squared Automatic Interaction Detection (CHAID) decision tree model was further developed and validated using 10-fold cross-validation, yielding a cross-validation risk value of 0.003, which was comparable to the resubstitution risk value (0.002), indicating good generalization ability and low risk of overfitting. A novel five-dimensional overlapping analysis identified RFW as the core evaluation indicator, with RDW, TFW and R/S as key auxiliary indicators. Conclusion This study delivers a standardized, reproducible technical framework for large-scale screening of saline-alkali-tolerant soybean germplasms. It facilitates global saline-alkali land utilization, accelerates worldwide soybean stress-tolerance breeding, and provides a transferable paradigm for stress tolerance evaluation in other major crops.
Abstract Purpose Timely, accurate, and field-scale crop yield mapping is essential for precision crop management, yet most existing studies rely on late- or full-season data, limiting in-season decision-making. This study aims to develop a stage-aware earliest possible corn yield mapping framework that balances early data availability, predictive accuracy, and spatial fidelity by integrating Sentinel-2 imagery, vegetation indices (VIs), and accumulated growing degree days (AGDD). Methods Corn yield data were collected from a commercial farm over three growing seasons (2018–2020). The final modeling dataset included 51,794 Sentinel-2 10 m aggregated yield samples across three seasons: 2018 ( n = 16400), 2019 ( n = 18534), and 2020 ( n = 16860). Sentinel-2 raw spectral bands and derived VIs were organized for V4, V6, R1, R5, and R6 growth stages based on AGDD- and DAP-defined corn growth-stage windows, also confirmed visually on-ground, allowing observations from different planting and harvest dates across the three growing seasons to be aligned by crop developmental stage rather than calendar date. A stage-wise Pearson correlation and frequency-based selection identified informative and non-redundant VI subsets across crop development. Four machine learning models: Random Forest (RF), XGBoost (XGB), k-Nearest Neighbors (kNN), and a Neural Network (NNET; multi-layer perceptron) were trained using 13 input configurations, including individual growth stages and multi-stage combinations capturing phenological progression. Models were tuned via randomized search, trained on 2018–2019 data, and independently validated on the 2020 season. Yield predictions were mapped directly at 10 m Sentinel-2-pixel resolution without spatial interpolation to preserve fine-scale variability. Results Model performance was strongly influenced by phenological stage selection. Among single-stage inputs, R1 was the earliest stage where reliable yield mapping could be availed (RF: R² = 0.56, RMSE = 29.50%). While combining V6 with R1 stage inputs substantially improved predictive performance (RF: R² = 0.70, RMSE = 24.13%) for the yield mapping at the R1 stage, where V6-stage signals provided complementary yield-related information. The full-season combination (V4 + V6 + R1 + R5 + R6) produced the highest accuracy (RF: R² = 0.72, RMSE = 23.28%) but would be less suitable for early in-season decision-making. Early- or late-stage-only inputs (V4, R5, R6) showed weaker and less stable cross-year performance. Among algorithms, RF consistently generalized best to the independent 2020 dataset and is recommended for operational use. XGB showed strong training performance but reduced cross-year stability, kNN yielded moderate accuracy, and NNET achieved accuracy comparable to RF while closely reproducing observed spatial patterns such as center-pivot effects and edge gradients. Direct 10 m mapping preserved yield heterogeneity and avoided smoothing artifacts common in interpolation-based approaches. Conclusion Stage-aware feature selection and phenology-informed input combinations are critical for balancing yield prediction timeliness and accuracy. For operational in-season yield mapping, the RF model using the V6 + R1 stage combination provides a practical and reliable solution, enabling early, accurate, and spatially detailed yield estimates with robust cross-year performance. This study presents a deployable framework for integrating satellite time series and weather data into conventional (non-sequential) machine learning models to support proactive, within-season decision-making in precision agriculture.
Conventional breeding for ideotypes in target environments remains challenging due to genotype-by-environment interactions and the genetic complexity of key agronomic traits. Traditional multi-environment field trials are costly and time-consuming, limiting rapid genetic gain. These challenges highlight the need for digital tools to support rice breeding. However, two major approaches, genomic prediction (GP) and gene-based crop models (GBCMs), have distinct advantages. In this study, a dataset derived from a natural rice population comprising 210 genotypes, genotyped with about 650,000 markers, rice dry matter, and yield across three environments, was used to develop two genomic prediction models, genomic best linear unbiased prediction (GBLUP) and a convolutional neural network (CNN), together with a gene-based crop modeling framework. The effectiveness of these models in predicting rice traits and assisting in breeding selection was subsequently evaluated. Prediction results indicated that biomass and yield could be effectively predicted by all models, with Normalized Root Mean Square Error (NRMSE) values ranging from 10.60% to 18.59% and 9.93% to 18.19%, respectively. In terms of predictive accuracy, parameter-based crop models achieved the highest predictive accuracy, although it was confined to theoretical simulations. This was followed by the GBCM and CNN, whereas the GBLUP exhibited the lowest performance. Furthermore, GGE biplot analysis revealed the predictions of the GBCM aligned more closely with field observations than those of the CNN, emphasizing the potential of GBCM as a practical surrogate for digital breeding. These results provide valuable insights into modeling genotype-by-environment interactions and support the development of data-informed breeding strategies for future rice improvement.
Accurate long-term monitoring of winter wheat phenology is important for crop growth assessment and irrigation management, but 30 m Landsat-based phenology retrieval remains challenging because of sparse observations, cloud contamination and sensor differences. This study developed and evaluated an integrated Landsat-based workflow for monitoring winter wheat phenology in the People’s Victory Canal (PVC) Irrigation Area of northern Henan and the Alar Irrigation Area of southern Xinjiang from 2000 to 2024. Winter wheat areas were mapped using temporally stacked NDVI/EVI features and a CART classifier. Vegetation-index trajectories were reconstructed using locally adjusted cubic-spline capping combined with Savitzky–Golay filtering, and green-up, jointing, heading and maturity were extracted using threshold- and derivative-based detection. The CART-based mapping achieved an overall accuracy of 89.51%, with higher accuracy in Alar (91.45%) than in PVC (84.35%). Compared with S-G-only, Whittaker and TIMESAT-like approaches, LACC + S-G reduced phenological-date errors, especially for green-up and maturity, with RMSE values within 3.1 d against agro-meteorological observations. Phenological stages generally occurred later in Alar than in PVC, and spatial autocorrelation confirmed significant clustering. Agro-meteorological analysis suggested stronger thermal associations in PVC and stronger moisture-related associations in Alar. These results provide practical 30 m phenological information for regional winter wheat monitoring and irrigation scheduling analysis.
Lodging is a major yield-limiting factor in soybean, but efficient large-scale phenotyping and genetic dissection of this complex trait remain challenging for breeding programs. To bridge this gap, this study developed an integrated, breeding-oriented framework that links UAV-based high-throughput phenotyping with candidate gene identification. Field experiments involving 741 diverse soybean genotypes were conducted over two years, with UAV remote sensing performed at key reproductive stages (from R5 to R7). We identified UAV-derived structural (relative plant height), textural (homogeneity, dissimilarity, correlation), and spectral (NDVI, EVI, NDRE) features as the most sensitive indices for retrieving lodging severity. The fusion of these complementary features, coupled with the XGBoost algorithm, achieved high classification accuracy (0.81–0.92) across genotypes, growth stages, and years. This reliable phenotyping pipeline enabled the precise selection of contrasting genotypes (lodging-resistant vs. lodging-prone) for transcriptomic analysis. Transcriptome sequencing revealed 13,447 differentially expressed genes, with significant enrichment in phenylpropanoid and starch–sucrose metabolic pathways. Moreover, the haplotype analysis within a natural population identified superior allelic variants of two candidate genes ( Glyma.19G249100 and Glyma.05G142200 ) significantly associated with soybean lodging resistance. This work can effectively bridge the gap between scalable field phenotyping and the discovery of functionally validated breeding targets, providing an efficient and translational framework to accelerate the development of lodging-resistant soybean varieties.
Canopy 3D architecture is a critical determinant of light interception, photosynthetic efficiency, and final yield in cotton, yet its rapid and accurate characterisation remains challenging in field conditions. To achieve efficient, non-destructive, and quantitative monitoring of the canopy structure of field-grown cotton, this study proposes a 3D structure-based technology stack for high-efficiency, low-cost, and high-precision phenotyping extraction. This stack directly addresses the technical bottlenecks of traditional 3D data acquisition, namely high cost, long processing time, and low operational efficiency, which have hindered large-scale application. We developed a pipeline that integrates a fast reconstruction algorithm with a scale-recovery mechanism using ground control points (GCPs), enabling the generation of true-scale 3D point clouds from UAV aerial images in a cost- and time-effective manner. Using only 141 UAV images and with a reconstruction time of approximately 20 min, we efficiently reconstructed high-quality, scale-accurate point clouds of two 5.5 m × 5.5 m cotton plots, significantly outperforming SfM-MVS and Instant-NGP in terms of both reconstruction efficiency and point cloud completeness. This method, whose current validation is confined to a single season, one growth stage, and two experimental plots, not only achieves a breakthrough by using fewer input images with high efficiency, but also ensures point cloud accuracy and completeness, showing strong potential for rapid field monitoring and real-time management. Based on the high-quality reconstructed point clouds, we further quantitatively evaluated canopy characteristics at harvest, analyzing the coefficient of variation of canopy height, porosity distribution, and canopy volume fraction. The core shortcomings and optimization strategies for the existing canopy structure were identified, providing scientific data support and practical technical references for precision cultivation management and mechanization-compatible planting in cotton.
Accurate estimation of the cotton seedling Leaf Area Index (LAI) is essential for yield prediction and precision crop management. Traditional manual methods are limited by low throughput, while two-dimensional remote sensing techniques often struggle with sparse canopy cover and soil background interference. Although three-dimensional LiDAR presents a promising alternative, existing deep learning approaches typically depend on costly fully supervised point-wise annotations. To address this challenge, this study proposes an end-to-end framework that integrates UAV-based LiDAR, weakly supervised segmentation, and physical parameter inversion. A weakly supervised network, termed CogNet, was developed—incorporating self-distillation and structure-aware label propagation—to achieve precise segmentation of cotton plants using only 10% sparse annotations. Following instance segmentation via Density-Based Spatial Clustering of Applications with Noise (DBSCAN), individual plant phenotypic traits were extracted. A nonlinear Extreme Gradient Boosting (XGBoost) model was then constructed to invert LAI by leveraging allometric relationships between 3D structural parameters and leaf area. Experimental results showed that CogNet achieved an Intersection over Union (IoU) of 85.34%, effectively mitigating overfitting to label noise and achieving performance competitive with the fully supervised RandLA-Net (82.13%). Notably, under the specific conditions of this cotton seedling dataset characterized by strong geometric priors, the weakly supervised model demonstrated enhanced robustness against annotation inconsistencies. The framework attained a plant detection rate of 96.2%, and the XGBoost model delivered high estimation accuracy (R² = 0.879, RMSE = 0.138). This study demonstrates that weakly supervised learning can substantially reduce annotation costs while maintaining model performance, providing an efficient and cost-effective solution for field-scale crop phenotyping.
Early detection of latent fungal decay caused by Penicillium italicum(P. italicum) and Penicillium digitatum(P. digitatum) remains challenging due to the absence of visible symptoms. In this study, a Vis-NIR hyperspectral imaging framework was developed to characterize early biochemical alterations in navel oranges. To address sample scarcity, a generative modeling approach (WGAN-GP) was employed to capture the intrinsic physiological variability of infected tissues. The successive projections algorithm (SPA) identified 20 key wavelengths associated with water redistribution (OH), carbohydrate depletion (CH), and chlorophyll degradation. These wavelengths were expanded into continuous ROI windows (W = 17), enabling integration of narrow-band pigment signals and broad-band absorptions related to water and carbohydrates via a multi-scale mixture-of-experts (MS-MoE) network. The framework achieved a classification accuracy of 97.10% and an F1-score of 0.9666. These results demonstrate that specific spectral absorption windows can serve as reliable, chemically interpretable spectral biomarkers for detecting early pathological changes in citrus fruit.
Sun-induced chlorophyll fluorescence (SIF) is an effective proxy for vegetation photosynthesis, but tower-based retrieval suffers from atmospheric path interference under humid and variable conditions. We present a DOAS-based SIF retrieval algorithm that operates in Fraunhofer lines (680–686 nm, 745–758 nm) and water vapour-sensitive bands (717–727 nm). It constructs an adaptive reference spectrum from SCOPE simulations and PCA and incorporates H2O absorption cross-sections into the fitting process for active atmospheric correction. The algorithm is implemented in a dedicated tower-based system integrating a 1° scanning gimbal with a high-resolution spectrometer. Validation with simulated and field data demonstrates the following: (1) the algorithm retrieves SIF with high fidelity (correlation coefficients >0.9 across all windows); (2) it exhibits lower water-vapour sensitivity and greater cloudy-sky stability than FLD, 3FLD, and SFM, achieving the lowest coefficient of variation (CV = 0.356); (3) over a complete wheat–rice rotation, the retrieved SIF tracks crop growth and phenological stages. This work provides a reliable solution for automated, high-precision tower-based SIF observation under complex atmospheric conditions.
Live imaging of plant subcellular structures is key to deciphering the spatiotemporal bases of cellular processes, and their functional impact on growth and morphogenesis at various biological scales. Live imaging of plant cells essentially relies on expression of fluorescent markers labeling cells or subcellular structures of interest. Simultaneous multi-channel imaging of several markers is still not routine practice in plant cell biology, owing to issues linked to genetic or spectral compatibility of markers, differences in expression levels, silencing, toxicity, etc. Here we designed a three-color marker in Arabidopsis thaliana and Capsella rubella , enabling high-resolution live imaging of plant morphogenesis, including labeling of the cell membrane, the nucleus and the microtubule cytoskeleton. Detection of MT arrays involved the development of a MAP4-MBD-based microtubule marker optimized for plant cells. The three- color marker allows visualization of the three-dimensional organization and dynamics of plant microtubules within the intracellular space with unprecedented precision, in various organs including the root and shoot meristems, the leaf, anther, and gynoecium. Our results demonstrate the potential of such single-construct strategy for cell biology studies in plants.
Accurately quantifying the areas of necrotic lesions on plant leaves is essential for evaluating plant-pathogen interactions and disease resistance. Although digital image analysis methods using ImageJ are widely employed, they often require case-specific optimization and may not be readily applicable across different experimental conditions. Furthermore, many studies have used ImageJ for lesion measurement without providing methodological details, which limits reproducibility. Here, we present a simple, step-by-step ImageJ workflow for measuring irregular necrotic lesions using a standard personal computer and mouse. The procedure relies on manual lesion selection using the freehand selection tool, followed by Gaussian smoothing, binarization, and automated particle analysis to extract lesion area measurements. By balancing manual isolation with computational thresholding, this protocol eliminates the need for extensive parameter tuning. This approach provides an accessible, reliable alternative to time-consuming color thresholding methods, thereby improving transparency and reproducibility in lesion quantification. The workflow's reproducibility has been confirmed through both intra-user and inter-user analyses. Key features • Relies on simple manual freehand selection combined with minimal image processing, requiring only a standard computer and mouse without specialized software or advanced training. • Enables accurate quantification of irregular necrotic lesions in conditions where automated thresholding methods require time-consuming optimization. • Provides a fully detailed, reproducible ImageJ workflow addressing common gaps in published methods, facilitating direct implementation. • Demonstrates high reproducibility, validated by intra-user and inter-user statistical analyses, ensuring reliable lesion quantification regardless of the operator.
Accurate quantification of plant disease severity is essential for evaluating host-pathogen interactions and assessing the effectiveness of disease management strategies. Traditional visual scoring methods and manual estimation of infected tissue are widely used but are often subjective and prone to observer bias. Digital image analysis offers an objective alternative by enabling automated identification and quantification of symptomatic plant tissues based on color and spatial characteristics. Here, we present a MATLAB-based image processing protocol for differentiating diseased and healthy plant tissue from digital leaf images. The workflow involves acquisition of standardized leaf images, conversion of RGB images into hue-saturation-value (HSV) color space, segmentation of diseased tissue using defined HSV thresholds, refinement of the segmented mask through morphological operations, and extraction of the whole leaf area. The protocol then calculates the diseased area and total leaf area in pixels and computes the percentage of infected tissue. The method uses MATLAB together with the Image Processing Toolbox and can be implemented using simple scripts. This protocol enables rapid and reproducible quantification of disease severity in plant leaves exhibiting visually distinct symptoms such as necrotic lesions or blight patches. By minimizing observer bias and providing quantitative measurements of infected area, the protocol offers a practical and reproducible approach for plant disease phenotyping and evaluation of disease management strategies across diverse plant-pathogen systems where diseased tissues can be clearly distinguished from healthy tissues under reasonably controlled imaging conditions. Key features • A reproducible MATLAB-based workflow for separating diseased and healthy plant tissue using color-space segmentation. • Applicable to plant diseases where symptomatic tissue contrasts clearly with healthy tissue (necrosis, blight lesions, rot patches). • Requires digital leaf images, MATLAB, and the MATLAB Image Processing Toolbox for image processing and disease quantification. • Enables rapid calculation of diseased leaf area and disease severity using automated pixel-based quantification.
Reproduction assets foundThe protocol explicitly deposits its authors' MATLAB image-processing workflow (HSV segmentation, mask refinement, pixel-based disease quantification) in a public GitHub repository with README instructions and example images.Code · publicGitHub repository containing the MATLAB source code, README file with installation and execution instructions, and representative example image(s): https://github.com/pankajborahmajuli-source/Leaf-Disease-Detection-MATLAB-Code/blob/main/README.mdOpen asset ↗Leaf-Disease-Detection-MATLAB-Codehtml-lines:112-148Plant phenotyping relevance matchCrossref · checked 15 Sept 2026
Food safety globally is threatened by crop disease, which creates a major obstacle to yield losses, so there is an urgent need for rapid, precise, and large-scale diagnostic methods for all the global risks crops are exposed to from disease. While imaging sensors, as well as Artificial Intelligence (AI), have made great strides in recognising plant disease, most literature does not have a comprehensive analysis that combines methods, technology, and implementation. Therefore, a systematic literature review follows PRISMA methods; we review 61 excellent studies published within the last five years that outline the advancement of imaging modalities (Red, Green, Blue (RGB), multispectral/ hyperspectral, thermal), deep learning architectures, augmentation of data, explanation methods and IoT (Internet of Things)-edge-cloud for managing intelligent agriculture. These modern AI-based systems (AI systems) have consistently produced accurate results above 98%. However, there are problems with the generalisability (across hybrid plant species), robustness (when exposed to environmental stresses), and interpretability of the results presented to consumers. This review represents the first compilation of using imaging sensors, artificial intelligence models, Internet of Things architecture (IoT-edge), and robotics into one comprehensive framework for the detection of plant disease in the next generation. In addition, this review suggests future research directions, including lightweight edge-deployable models, multimodal sensor fusion, interpretable AI, larger validated datasets, and autonomous robotic systems for scalable and sustainable smart agriculture.
Fluorescence in situ hybridisation (FISH) is a valuable technique for visualising RNA molecules in their native cellular context. Still, its application in plant tissues is often limited by tissue autofluorescence and the lack of optimised protocols. Here, we developed and validated a simple, reproducible FISH workflow to detect Apple scar skin viroid (ASSVd) in cucumber. Systematic optimisation of probe chemistry, tissue selection, and sampling stage significantly improved assay sensitivity and reproducibility. The AZDye594-labelled antisense riboprobe produced higher signal-to-background ratios and lower background fluorescence than fluorescein-labelled probes, enabling reliable detection of ASSVd in vascular-associated tissues. The optimised workflow consistently detected ASSVd in both leaves and stems. High-resolution confocal imaging further revealed predominant nuclear accumulation of ASSVd RNA in infected cells. Together, this study establishes a sensitive and accessible FISH workflow for localisation of ASSVd in cucumber and provides a practical platform for investigating the spatial distribution of viroid and other plant RNA pathogens.
Andrew Trlica · Rachel L. Cook · Matthew J. Sumnall
Field / plotLiDAR / point cloudMultispectral / hyperspectralLeafMorphology / geometry measurementLeaf traits
Canopy Leaf Area Index (CLAI) is a stand attribute containing information on the real-time health and growth potential of managed pine plantations. Current remote sensing techniques for quantifying CLAI rely on simple linear models applied to satellite multispectral imagery, or on techniques based on light detection and ranging (LiDAR) data that are costly and less frequently collected. This study demonstrates a convolutional neural network (CNN) approach to retrieving CLAI from 10 m Sentinel-2 multispectral imagery with a model trained on gridded LiDAR-based CLAI estimates. We demonstrate large gains in accuracy with the CNN compared to traditional linear models based on vegetation indices (e.g., Simple Ratio), but also clear shortfalls in model skill when predicting “blind” in some spatial domains that were completely excluded during model training. Pixel-scale root mean squared error ranged from 0.34 to 0.64 by domain when exposed to CLAI training data from all available spatial domains, but rose to 0.58–1.74 when predicting without prior domain-specific training. Prediction accuracy was consistently lower when applied to completely unobserved USGS LiDAR-based CLAI estimates. Traditional linear models, in contrast, had the advantage of usually lower prediction error across unobserved spatial domains (0.43–1.98), but with lower maximum accuracy. These results demonstrate a potential route for deploying more complex models for LiDAR “mimicry”, e.g., between data acquisitions widely separated in time, but advocate for the development and use of more stable generalized approaches for use in unobserved managed pine stands.
Jiayuan Yang · Juntao Xiong · Mingyue Zhang · Baoxia Sun · Hongxing Peng
NeRF / 3D Gaussian SplattingFruit2D/3D reconstructionSegmentation
High-fidelity and lightweight 3D fruit models are a crucial foundation for phenotypic analysis and automated agricultural robotic operations. However, in complex agricultural scenarios characterized by varying illumination and foliage occlusion, existing 3D reconstruction methods struggle to balance reconstruction accuracy and model size, often generating massive redundant background primitives. To address this challenge, this paper proposes a novel framework for lightweight and high-fidelity 3DGS fruit reconstruction via geometric-semantic joint constraints. Specifically, the method first integrates depth priors and semantic information through a Depth-Guided Semantic Segmentation module to extract accurate target fruit masks. Next, it eliminates background noise points from the initial point cloud using a multi-view Reprojection Consistency Voting mechanism. Simultaneously, a Stochastic Background Regularized Hybrid Loss is introduced during the 3DGS training phase to decouple density and color optimization, thereby suppressing the regeneration of background Gaussian primitives. Experimental results on a multi-category fruit dataset demonstrate that while maintaining a high novel view synthesis quality (PSNR of 31.87 dB), our proposed method reduces the average model size from 230.42 MB to 62.66 MB (a 72.8% reduction), achieving robust, high-fidelity, and lightweight 3D fruit reconstruction.
Cell polarity and tip growth rely on the dynamic spatial organization of signaling and structural components. Quantitative characterization of these spatiotemporal dynamics is critical for understanding polarized cell growth, yet manual quantification is labor-intensive and existing computational tools often lack the flexibility and robustness needed to analyze molecular and structural dynamics in tip-growing cells. Tip Quantification (TipQuant) identifies the cell apex by detecting the site of maximum expansion and automatically quantifies fluorescence distribution along the plasma membrane and within the apical cytoplasm from live-cell imaging data, enabling analysis of the spatiotemporal dynamics of molecular and structural components in tip-growing cells. TipQuant accurately identified cell apices and quantified the spatiotemporal behavior of fluorescently labeled proteins and cellular structures in Arabidopsis thaliana pollen tubes and Fusarium graminearum hyphae, reproducing manual measurements while reducing user bias and improving efficiency, consistency, and analytical flexibility. The tool also revealed a strong positive correlation between rho-like GTPase from plants activity and apical Ca 2+ influx in Arabidopsis pollen tubes, demonstrating its utility for analyzing dynamic cellular processes. TipQuant is a robust analytical tool for quantifying spatiotemporal dynamics in tip-growing cells, providing a flexible alternative to manual image analysis and enabling studies of the molecular mechanisms underlying polarized growth.
Wheat grain number integrates fate of individual florets, strongly affected by environmental stress during sensitive stages like meiosis. Because development is asynchronous across tillers, spikelets, and florets, it is hard to distinguish stress tolerance from stress escape. Based on 2400 destructive measurements of spike and anther length together with non-destructive morphological measurements from 158 plants grown in four experiments in controlled-environment, we developed a framework to track individual floret developmental stages at plant level. We applied it in two case studies for connecting within-plant developmental asynchrony to reproductive success under favorable or heat conditions. All florets showed a common relative growth rate, producing additive delays across tillers (1-7 d), spikelets (1-5 d), and floret positions (1-6 d). This generated a developmental map for every floret based on external traits. Under control conditions, grain set probability at floret level combined both positional and developmental effects within a spike. Under heat stress, grain loss occurred only in florets at meiosis during the stress, allowing to quantify a true "stress response", while later florets escaped damage. This framework allows understanding and predicting floret development and linking it to grain set, clearly distinguishing timing effects from positional influences and separating tolerance from stress escape.
Manisha Das Chaity · Ramesh Bhatta · Byron Eng · Jan van Aardt
Field / plotPhotogrammetry / SfM / MVSLiDAR / point cloudMultispectral / hyperspectralWhole plant / canopy / plot / field2D/3D reconstruction
The Greater Cape Floristic Region (GCFR) in South Africa is a fire-prone biodiversity hotspot where high species richness, structural complexity, and small plant sizes (0.0001–4 m2) pose substantial challenges for remote sensing-based biodiversity assessment. Spectral similarity among species and the mismatch between plant size and sensor pixel dimensions limit the capacity of current and forthcoming spaceborne systems to resolve individual species and accurately detect plot-level diversity changes. We therefore developed a physics-based simulation framework that couples fynbos trait measurements with radiative transfer modeling in the DIRSIG (Digital Imaging and Remote Sensing Image Generation) environment towards quantifying information loss across spectral and spatial scales and to define theoretical limits for biodiversity monitoring. We constructed a three-dimensional virtual scene of post-fire fynbos communities in Grootbos Private Nature Reserve, integrating high-resolution imagery, terrestrial laser scanning (TLS), and structure-from-motion (SfM)-derived point clouds. Field measurements of mean diameter and percent cover were used to scale vegetation models and constrain species abundance. We distributed plant instances using a blue noise sampling algorithm, guided by density maps derived from unmanned aerial system (UAS) imagery. Species-specific optical properties were parameterized using field-measured reflectance data and the PROSPECT radiative transfer model, while terrain structure was derived from SfM-based digital terrain models. The integrated scene was used to simulate multispectral (DJI Mavic 3 MSI), hyperspectral (AVIRIS-NG), and light detection and ranging (LiDAR) observations. Agreement between simulated outputs were evaluated against corresponding field-acquired datasets using spectral signatures and vegetation indices. This framework enables systematic assessment of sensor specification effects on spectral biodiversity metrics and provides a pathway for evaluating theoretical limits of species discrimination across airborne and satellite platforms.
Introduction In a complex orchard environment, canopy apples are obscured by various factors, making it hard for apple harvesting robots to accurately determine which apples can be directly harvested. Furthermore, the complex obstruction leads to difficulties in identifying keypoints on the apples and caculating the axis direction, directly affecting the robot's determination of grasping positions. Methods To solve these issues, a two-dimensional (2D) axis estimation method for pickable canopy apples based on a YOLO cascade network was proposed. Firstly, the study introduced SPDConv for lossless downsampling and adopts dynamic upsampling to improve segmentation boundary accuracy, constructing an instance segmentation network named the YOLO-SD model to select pickable apples according to different occlusion conditions and growth states. Secondly, the geometric center of the mask image was located using its minimum enclosing circle, and the region of interest was extracted through morphological dilation. Then, by integrating the RFAConv, SCSA attention, and MBConv modules, a keypoint detection network YOLO-RSM was constructed to extract keypoints of pickable apples. Finally, a 2D axis construction strategy was proposed, which adaptively constructs the growth axis based on the visibility of keypoints. Results Experimental results show that the overall average accuracy mAP50 of the YOLO-SD model for apple segmentation reached 95.2%, and the parameter quantity was reduced to 2.47 M. The average accuracy of the YOLO-RSM model for keypoint detection has reached 90.3%, which is 2.4%, 2.6%, and 4.7% higher than that of the YOLOv8n, YOLO11n, and YOLO12n models respectively. The 2D axis estimation algorithm has an average axis angular error of 7.23° ± 16.73°, and an axis estimation accuracy of 92.68%. Discussion The proposed method can achieve high-precision canopy apple segmentation, keypoint detection, and 2D axis estimation, thus offering technical support for the picking operations of apple harvesting robots.
Plant secondary metabolites are mainly synthesized and stored in secretory tissues. Secretory canal development has been mainly characterized in Apiaceae. The secretory canals of Peucedanum praeruptorum contain pharmacologically active coumarins, but their organ-specific distribution and developmental dynamics remain poorly understood. This study integrated light microscopy (LM), transmission electron microscopy (TEM), X-ray microcomputed tomography (µ-CT), and high-performance liquid chromatography (HPLC) to investigate canal development, distribution, ultrastructure, 3D architecture, and coumarin accumulation in P. praeruptorum roots. Histological analysis showed that canals adjacent to the periderm originate from pericycle cells, whereas those in secondary phloem arise from parenchyma differentiation; both develop schizogenously. Canal quantity and dimensions varied temporally. Canals located in phloem showed density increasing toward the cambial zone, where cross-sectional areas were smaller. The canal density index increased from September to November, peaking on November 15, then declined. HPLC revealed dynamic accumulation of five major coumarins: content increased from September, peaked on November 15, then gradually decreased. TEM showed that epithelial cells surrounding the canal lumen were rich in Golgi, ER, mitochondria, plastids, starch grains, and osmiophilic droplets. µ-CT volumetric analysis and segmentation generated detailed 3D models, revealing spatial organization and enabling size-based grouping of canals (1000-3000 μm). These dimensional characteristics aligned with developmental progression. This study characterizes the ontogeny, distribution, ultrastructure, and 3D architecture of secretory canals, providing a structural foundation for investigating correlations between secretory tissues and compound synthesis.
Hazra M, Crowther A, McInerney F, Strömberg CAE, Fabillo M.
LeafSeed / grainClassification
Background and aims Phytolith analysis is widely applied in palaeoecological and archaeological research, but its interpretive strength depends on the availability of robust modern reference collections. This study expands the modern Australian phytolith reference collection by analysing 42 native plant species representing 24 families and 37 genera with emphasis on silicification patterns across major growth forms, including forbs, shrubs, trees, and C3 grasses. Methods Phytoliths were extracted from available plant parts, including leaves, stems, flowers, seeds, seed pods, cones, and roots, depending on sample availability. Morphotypes were identified following ICPN 2.0, with grass silica short cell phytoliths (GSSCPs) further classified by shape and size to examine subfamily-level patterns. Phytolith morphotype percentage data were analysed using Hellinger transformation, PerMANOVA, PCA, LDA, and hierarchical clustering to assess compositional differences among plant growth forms and grass subfamilies. Key results Phytolith production varied strongly among growth forms and plant parts. Grasses were abundant producers, whereas most forbs, shrubs, and trees were trace producers or non-producers. Leaves were the most consistent source of phytoliths, while seeds and seed pods were predominantly non-producers. Grass silica short-cell phytolith (GSSCP) morphotypes showed clear subfamily-level differentiation. Pooideae produced Rondel morphotypes. Danthonoideae produced Rondel as well as wide Bilobate types. Panicoideae and Oryzoideae exhibited a pronounced Bilobate signature, commonly associated with Polylobate and Cross forms. Non-grass taxa (woody, shrubs, and forbs) were dominated by Spheroids, Tracheary elements, Epidermal, and Polygonal sheets and other non-diagnostic forms. Phytolith assemblages differed significantly among plant families, with Poaceae uniquely producing GSSCPs, while non-grass families showed greater overlap in assemblage composition. Conclusion By expanding taxonomic and anatomical coverage, this study strengthens the capabilities of phytoliths in the reconstruction of grasslands and in general paleo vegetation in Australia, especially where other proxies such as pollen are limited.
Reproduction assets foundThe authors explicitly state that the R scripts used for data analysis and figure generation are publicly available on their GitHub repository, which directly reproduces this paper's phytolith statistical analyses (PCA, LDA, PerMANOVA, clustering, plots). Supplementary data files contain the paper's measurements but noCode · publicntification of all plant specimens collected for this
study.
A
14 FUNDING
M
15 Funding for this study was provided by ARC Discovery grant DP210100508 and a Ph.D.
D
16 fellowship (UQGSS) to MH.
TE
17 DATA AVAILABILITY
18 The R scripts used for data analysis and figure generation are publicly available on GitHub
EP
19 repository: https://github.com/Manoshi-sporo/Australian-Phytolith-Reference-Collection.
20 CONFLICTS OF INTEREST
CC
21 The authors declare no competing financial or commercial interests.
A
22 AUTHOR CONTRIBUTIONS
23 MH: writing original draft, conceptualization, software, investigation. AC: Supervision, Writing -
24 Review and editing, FM: Supervision, Writing-Review and editing.Open asset ↗Manoshi-sporo/Australian-Phytolith-Reference-Collectionpdf-layout-page:34 lines:1-87Plant phenotyping relevance matchCrossref · checked 5 Sept 2026
Unmanned aerial vehicle (UAV)-based multispectral imaging enables flexible, non-destructive crop monitoring. Although UAV imagery offers much higher spatial resolution than satellite platforms, its effective spatial detail at typical operational flight altitudes can still be insufficient for plant-level interpretation and fine canopy structure, which can reduce vegetation-index reliability. Most super-resolution (SR) research targets RGB or satellite imagery and emphasizes perceptual or pixel-wise quality, leaving the spectral fidelity of reconstructed UAV multispectral imagery under-examined. This study benchmarked an SR evaluation framework for UAV-based five-band crop imagery (Blue, Green, Red, Red-edge, and near-infrared) using the open-source AI Hub cabbage dataset, with low-resolution inputs generated by controlled downsampling at ×2, ×3, and ×4. Nine methods (bicubic, SRCNN, EDSR, RCAN, SwinIR-based, ESRGAN-based, HAT-based, DAT-based, and DRCT-based SR) were compared under joint five-channel and band-wise reconstruction on 2170 test scenes using image-quality, spectral-angle, vegetation-index (NDVI, GNDVI, NDRE), band-wise, and efficiency metrics. EDSR and RCAN gave the most balanced performance. At ×4, band-wise reconstruction was strongest for per-band spatial fidelity, where EDSR reduced RMSE by 12.6%, and RCAN lowered near-infrared RMSE by about 21% relative to bicubic, whereas joint reconstruction with its spectral-angle and vegetation-index losses best preserved spectral relationships (spectral angle and vegetation-index errors). Learning-based gains were clearest at ×4. The recently proposed HAT-based, DAT-based, and DRCT-based attention models achieved the strongest pixel-wise RMSE and PSNR but did not surpass EDSR or RCAN on spectral angle or vegetation-index preservation under the equalized training budget. These results indicate that UAV multispectral SR should be assessed by spatial fidelity together with spectral consistency and agricultural index reliability.
Abstract In real field scenarios in agriculture, automatic segmentation of plant diseases is an important technique for precision farming. However, it remains exceptionally challenging due to blurred lesions, complex morphological structures, irregular backgrounds, and severe class imbalance. While traditional convo lutional networks struggle to capture long-range semantic context and standard vision transformers fail to preserve sharp localized boundaries, this paper proposes an efficient, attention-gated hybrid framework optimized for field deployment. Our architecture leverages a hierarchical Mix Transformer (MiT-B2) encoder stream integrated with an Atrous Spatial Pyramid Pooling (ASPP) scale-space context bridge and a custom Cross-Scale Multimodal Attention Gate (CMAG) to isolate discriminative disease markers selectively. Evaluated on the highly challenging and unbalanced PlantSeg dataset, our framework achieves competitive mean Intersection over Union (mIoU) of 66.57% and an F1-score of 79.93%, while maintaining a highly compact parameter footprint of only 30.37 M. Experimental evaluations demonstrate that the proposed system establishes a new performance milestone, outperforming current competitive architectures and proving highly viable for resource-constrained edge devices. To further enhance out-of-distribution stability, we outline future directions to extend our top-performing candidate variants into a Level 1 meta-stacking ensemble optimized via few-shot learning and partial backbone fine-tuning.
To address the challenges posed by large lesion-scale variations, weak boundary cues, and high inter-class similarity in rice leaf disease detection, this study proposes an improved RT-DETRv2-R50-based rice disease detection model, termed CED-RTDETR. First, a Lightweight Contour-Guided Aggregation Backbone (LCGA-Backbone) is constructed. In the PResNet residual blocks, an InceptionDWConv2d-based direction-aware depthwise separable spatial mixing strategy is introduced to capture local, horizontal, and vertical lesion texture patterns with low computational overhead. Meanwhile, a Contour-guided Efficient Global Aggregation Block (CEGA Block) is embedded after the outputs of the C3, C4, and C5 stages. Through contour-difference enhancement, channel shuffle, group-wise efficient global aggregation, and bottleneck channel mixing, the proposed block strengthens the representation of the boundaries of small lesions, weak textures, and contextual semantics. Second, a Multi-scale Evidence Interaction Fusion Neck (MEIF-Neck) is used to perform multi-scale evidence interaction and salient-region competition after cross-scale feature concatenation. A lightweight feature reconstruction process is further implemented using Spatial-Evidence RepNCSPELAN (SE-RepNCSPELAN), which is developed from a YOLO-style feature fusion structure. Finally, a Direction-Amplitude Decoupled Deformable Attention (DAD-DA) mechanism is introduced to decompose sampling offsets into direction rotation residuals and radius gains while incorporating an aspect-ratio-aware geometric compression-restoration strategy, thereby improving the geometric stability of sampling locations during the decoding stage. Experimental results on the constructed rice leaf disease dataset show that CED-RTDETR achieves AP, AP50, and AP75 values of 29.5%, 75.2%, and 17.6%, respectively, outperforming RT-DETRv2-R50 by 4.1, 4.9, and 3.8 percentage points. To further evaluate the model on an additional public benchmark dataset, experiments were also conducted on the public Rice Disease Dataset. On this dataset, CED-RTDETR achieves AP, AP50, and AP75 values of 34.1%, 77.1%, and 23.2%, respectively, improving upon RT-DETRv2-R50 by 4.3, 5.6, and 3.4 percentage points. These results indicate that the proposed method achieves consistent overall performance improvements on both the constructed dataset and the public benchmark dataset.
Functional-structural plant models simulate plant responses to environmental conditions, but their development and evaluation are often limited by the lack of datasets combining detailed architectural and physiological measurements. Here, we present a comprehensive dataset acquired from four oil palm plants ( Elaeis guinnensis) grown under controlled and contrasting climate scenarios. The dataset includes (i) three-dimensional reconstructions of plant architecture derived from terrestrial lidar point clouds, (ii) leaf-level gas exchange measurements used to parameterize photosynthesis and stomatal conductance models, and (iii) continuous plant-scale measurements of CO 2 and H 2 O fluxes obtained in a microcosm under precisely monitored and manipulated environmental conditions (light, temperature, humidity, and CO 2 concentration) across height climate scenarios. By combining detailed structural data with physiological measurements at both leaf and whole-plant scales, this database has been designed to build and evaluate digital twins (or shadows) of plants functioning under controlled conditions. It provides a valuable resource for calibrating biophysical models (light interception and photosynthesis), benchmarking model predictions across scales, and investigating the consistency between leaf-level parameterization and plant-level fluxes. All data and processing workflows are openly available, facilitating reuse for model development, evaluation, and intercomparison in plant and crop modelling communities.
Plant diseases remain a threat to global agricultural productivity, food security and livelihoods, especially in developing countries where the availability of experts in agriculture is still limited. The recent progress in AI, particularly deep learning and computer vision, has ushered in new possibilities for automated plant disease diagnosis, especially for plant image-based systems. This paper provides a systematic review of the deep learning methods employed for plant disease diagnosis, highlighting CNN-based methods, the application of transfer learning, explainable AI (XAI) methods and deployment issues. In the framework of PRISMA 2020, the relevant peer reviewed literature from 2016 to 2025 was systematically identified, screened and analysed on the most important academic databases. The review compared some of the most popular architectures such as GoogLeNet, DenseNet-121, MobileNetV2, EfficientNet, Attention-CNNs and Vision Transformers. Results showed very high classification accuracy in controlled lab conditions with DenseNet-121 achieving ~99.75% accuracy with good computational efficiency. But it also revealed a big gap between the lab and the field, mainly due to environmental variations, domain shifts, and dependence on datasets. Some innovative and emerging technologies like explainable AI, hyperspectral imaging, few-shot learning, and lightweight mobile architectures showed promise of enhancing the interpretability, early detection of disease, and the use of smart phones in low-resource agricultural settings. In conclusion, the study suggests that in order to be implementable in the field, future intelligent agricultural diagnosis systems must be able to balance predictive accuracy, explainability, computational efficiency and field adaptability. The results enrich the existing knowledge on precision agriculture and serve as useful information for researchers, agricultural technologists, and policymakers working on the creation of AI-based systems for crop protection.
Amanda Ewen · Rodrigo Godoy Mendez · Karar Al-Shanoon · Dawn Omoluabi · Anjana Samarasinghe · M. Alejandra Oviedo-Ludena · Karina Chimbo Huatatoca · Kara Glor · Keiko Nabetani · H. R. Kutcher · Lipu Wang · Ian Stavness · Lingling Jin
Reliable and objective phenotyping is essential for plant breeding programs to characterize genetic variation and accelerate crop improvement. Conventional disease assessment relies on expert visual scoring, which is labor-intensive, subjective, and prone to inter- and intra-rater variability. Although image-based phenotyping methods have been proposed, many require manual intervention, specialized imaging setups, or single time-point measurements, limiting their ability to capture disease progression over time. Here, we present a pipeline for longitudinal plant disease phenotyping that quantifies wheat stripe rust and leaf rust progression from time-series images. The pipeline performs semi-automated leaf and automated pustule segmentation from images acquired in situ , enabling objective disease severity estimation with minimal user intervention and without requiring solid backgrounds or manual leaf manipulation or detachment. By extracting temporal traits, including disease severity trajectories and standardized area under the disease progress curve, the method provides a comprehensive characterization of disease development throughout infection. Association between automated and expert assessments was moderate for stripe rust ( R 2 = 0.58) and strong for leaf rust ( R 2 = 0.85), while expert inter-rater reliability was moderate for both diseases (ICC = 0.675 and 0.800, respectively). The proposed approach establishes a scalable and reproducible framework for longitudinal disease phenotyping in controlled environments, with broad applications in disease resistance screening and crop breeding.
Reproduction assets foundThe paper's Code and Data Availability section explicitly states that software and datasets (the phenotyping pipeline and imaging datasets) are publicly available at the authors' GitHub repository and project website, both of which are in the allowed URL list.Code · publicSoftware and datasets are available at:
https://github.com/USask-BINFO/greenskeye_analysis and https://greenskeye.usask.ca/speedbreeding/ .Open asset ↗USask-BINFO/greenskeye_analysislines:195-225Dataset · publicSoftware and datasets are available at:
https://github.com/USask-BINFO/greenskeye_analysis and https://greenskeye.usask.ca/speedbreeding/ .Open asset ↗lines:195-225Plant phenotyping relevance matchOpenAlex · Crossref · checked 5 Sept 2026
Plant diseases lead to substantial yield losses and pose a persistent threat to global food security, creating an urgent demand for high-throughput, accurate, scalable, and non-destructive disease-monitoring approaches. Remote sensing has emerged as a powerful tool, yet progress in plant disease detection remains fragmented across various disciplines, tasks, sensing methods, and data modalities. This review introduces a hex-view perspective to synthesise remote-sensing-based plant disease detection within a cohesive conceptual framework. Instead of treating sensing technologies, algorithms, and datasets independently, the hex-view incorporates six interconnected dimensions that jointly capture how biological processes, the measurement scale, and data characteristics constrain disease detectability, including when detection is possible and how reliably it can be achieved. The hex-view framework comprises six interconnected dimensions and forms an integrated framework called BTSCAD: (1) Biology (B): plant–pathogen interactions constituting the biological foundation of disease development and expression. (2) Task (T): the diverse disease-detection tasks and their corresponding research objectives. (3) Sensor (S): the sensing modalities that define the data acquisition type and richness of captured information. (4) Condition (C): the environmental conditions, sensing platforms, and spatial scales that shape disease observations and bridge controlled experiments and real-world deployment across leaf, canopy, plot, and regional scales. (5) Algorithm (A): the classical and state-of-the-art data-analysis algorithms used to extract disease-related information from sensor data. (6) Dataset (D): the data sources that underpin model development, evaluation, and generalisability. The hex-view perspective provides a clear framework for interpreting previous research and identifying future research directions. This review lays a structured foundation for developing robust, interpretable, and transferable disease-detection systems, supporting advancements in precision agriculture, high-throughput phenotyping, and sustainable crop production.
Localized soil-moisture deficits, that is, irregular sub-field patches where crops experience water stress well before visible wilting, are a leading cause of yield variability in row-crop agriculture. These zones are difficult to detect at the spatial resolution and revisit frequency required for timely irrigation response. This paper presents a reinforcement-learningguided autonomous quadrotor unmanned aerial vehicle (UAV) platform that fuses onboard Visual Simultaneous Localization and Mapping (Visual SLAM) with a pushbroom hyperspectral imaging payload to construct georeferenced, canopy-registered maps of a Crop Water-Stress Index (CWSI) in near real time. Rather than flying a fixed lawnmower survey, the platform is guided by an adaptive-sampling policy trained with Proximal Policy Optimization (PPO) that reallocates flight time and sensor dwell toward regions of emerging water stress as evidence accumulates mid-flight. We present the complete engineering pipeline: airframe and sensor design, a keyframe-based Visual SLAM front and back end that provides centimeter-scale geolocation without continuous reliance on Real-Time Kinematic (RTK) GNSS lock, a hyperspectral preprocessing and spectralindex chain (NDVI, NDRE, NDWI/NDMI) used to derive CWSI through a learned regression, the partially observable Markov Decision Process (POMDP) formulation and reward shaping used to train the sampling policy, and the fused system architecture tying these subsystems together. In simulated field trials over a 0.8-hectare test plot, the reinforcement-learning-guided policy achieved a 92% water-stress-zone detection rate versus 61% for a fixed-grid baseline, while reducing mission flight time by approximately 32%. We further report an ablation study isolating the contribution of SLAM-derived canopy structure to CWSI accuracy, a sensitivity analysis across field complexity, and a full error budget for the fused pipeline. We close with a discussion of validation limitations, broader scientific and agricultural impact, and a roadmap toward multi-UAV fleet deployment for whole-farm monitoring
Abstract Satellite-based prediction of grain protein concentration (GPC) in wheat typically composites spectral observations over fixed calendar windows, implicitly assuming phenological synchrony across fields. We present a systematic evaluation of whether aligning multi-source remote sensing time series to field-specific, spectral-peak-relative windows improves field-level GPC prediction, for a quality trait whose physiology, senescence-linked nitrogen remobilization, contrasts with the season-integrating behavior of yield. Integrating Sentinel-2 imagery (31 vegetation indices, 10 spectral bands), ERA5-Land reanalysis, gSSURGO soil properties, and USGS 3DEP topography across 228 commercial winter wheat fields in western Kansas (2024–2025), we compared six temporal strategies (peakrelative vs. calendar × monthly, biweekly, growth-stage) using three ensemble tree models under nested cross-validation with Boruta feature selection. A single 30-day post-peak window (peak + [16,45] days) was the top-performing and most consistently selected window, chosen in 4 of 5 outer folds, reproducing prior accuracy under random cross-validation (R2 ≈ 0.28); though its advantage over the best calendar window was not statistically significant (paired bootstrap p = 0.08). Under leave-county spatial cross-validation, however, this skill did not transfer across counties (Sentinel-2–only R2 ≈ 0.01; per-county median R 2 = −0.23), indicating the satellite signal supports within-region interpolation but not spatial extrapolation to unseen counties; ablation shows that neither the spectral nor the static features transfer across counties on their own, and the residual crosscounty skill emerges only from their combination. A near-real-time application at ∼3 weeks before harvest retains most within-region skill at a modest accuracy cost. The results delineate where spectral-peak-relative alignment helps, concentrating a senescence-linked signal within region, and where it does not, providing an honest operational baseline for satellite-based grain-quality monitoring.
Reproduction assets foundThe preprint explicitly releases the authors' analysis code (data-acquisition pipeline, feature engineering, cross-validation/modeling, figure scripts) at a public GitHub repository, and a de-identified field-level GPC dataset released alongside the code repository. Both are paper-specific, public, and actionable. The Code · publicthe figure-generation scripts is available at https://github.com/Ciampitti-Lab/Open asset ↗Ciampitti-Labpdf-page:48 lines:1-55Dataset · publica de-identified version of the dataset is released alongside the code repositoryOpen asset ↗pdf-page:48 lines:1-55Code / dataset availability confirmedEurope PMC · checked 5 Sept 2026
Introduction Accurate classification of unsound wheat kernels is important for automated grain quality assessment, but improved recognition performance often comes at the cost of increased model complexity. Methods This study presents ECA-ModNet, a parameter-efficient convolutional network derived from EfficientNetV2-S. The architecture replaces two early-stage Fused-MBConv blocks with Mod-FusedMBConv blocks to introduce input-dependent local contextual modulation and replaces the squeeze-and-excitation modules in later stages with efficient channel attention to model local cross-channel interactions using fewer attention-related parameters. Experiments were conducted on the seven-class G600 wheat subset of the GrainSpace dataset. Results Across three independent runs, ECA-ModNet achieved an accuracy of 90.17 ± 0.21% and a macro-F1 score of 90.23 ± 0.21%, improving upon EfficientNetV2-S by 3.80 and 3.84 percentage points, respectively. The parameter count decreased from 20.19M to 16.49M, while FLOPs increased marginally from 2.90G to 2.95G. ECA-ModNet achieved accuracy statistically comparable to that of ConvNeXt-Tiny and InceptionNeXt-T while using substantially fewer parameters, and obtained 3.03-4.55 percentage points higher mean accuracy than six lightweight baselines. Discussion Ablation experiments identified two Stage 1 Mod-FusedMBConv blocks with a 3×3 context kernel as the configuration with the highest mean accuracy among those evaluated. These results indicate that ECA-ModNet offers a favorable accuracy-parameter trade-off for image-based classification of unsound wheat kernels.
Reproduction assets foundThe paper analyzes the public GrainSpace dataset (G600 seven-class unsound wheat kernel subset) and provides an explicit data availability statement with a public GitHub URL. No author analysis code or trained model deposit is stated.Dataset · publicPublicly available datasets were analyzed in this study. This data can be found here: https://github.com/hellodfan/GrainSpace .Open asset ↗hellodfan/GrainSpacelines:1035-1076Plant phenotyping relevance matchCrossref · checked 5 Sept 2026
Abstract Visual assessments of growing forest nursery plants are time-consuming and often result in a lack of information at a physiological level. There exists a need for health screening in nurseries, that is fast and efficient, to improve overall health monitoring and nursery productivity. Rapid handheld sensors such as rapid thermal devices, leaf porometers and moisture meters, can provide regular information at a physiological level, that can improve the understanding of the impact of stress on young plant cuttings and their decline in health over time. This paper evaluates the utility and reliability of contemporary sensor technologies, to operationally monitor stress phases in juvenile forest plant cuttings during progressive moisture (dry-down) conditions. Furthermore, to assess whether thermal sensors could be used as an indicator, in conjunction with other variables such as soil water content or stomatal conductance, is needed operationally for fast screening during limited planting windows. Near Infra-Red Analysis (NiRA) data was collected to understand detailed plant functions at a finer reflectance level. A relationship was found where the increase in thermal signals reflects a depletion of water content, resulting in an eventual decline in stomatal conductance and, ultimately, plant mortality. Several algorithms were used in a preliminary test, using RapidMiner software, to discriminate between the four phases of plant health decline using physiological variables and NiRA data. Both Gradient Boosting Trees (GBT) and Deep Learning (DL) showed the best performances, achieving favourable accuracies of 96.8% and 91.2% without NiRA data, 84.6% and 88.2% with NiRA data, with shorter training times. Using thermal technology weighted amongst the highest of the best performing variables using GBT, the utility and accuracy showed good discrimination between the stages of plant decline and is encouraged for future research in this field.
Mateo Pastrana · C. Velilla · Nelson Mattie · Alfonso Gomez · Sergio Molina
Field / plotLiDAR / point cloudWhole plant / canopy / plot / field2D/3D reconstructionYield / biomass estimationBiomass / plant weight
Reliable aboveground biomass (AGB) estimates for woody crops are essential for carbon accounting and for Measurement, Reporting and Verification (MRV) frameworks. However, it remains unclear how LiDAR modality and sampling geometry influence plot-scale and tree-scale AGB predictions in intensively managed Mediterranean orchards. In this study, we benchmarked four LiDAR modalities, namely open national airborne laser scanning from the Spanish National Aerial Orthophotography Plan (PNOA/ALS), a dedicated Riegl airborne laser scanner (ALS), unmanned laser scanning (ULS) and mobile laser scanning (MLS), across three woody-crop sites in Córdoba (southern Spain): IFAPA, Doña María, and Villaseca. Plot-level LiDAR metrics (mean height, 95th height percentile, maximum height, and canopy-cover proxies) were extracted from normalized point clouds and related to field AGB using Random Forest and XGBoost regression models, together with an ensemble predictor, under an 80/20 train–test split. In parallel, TreeQSM-based Quantitative Structure Models (QSMs) were evaluated as an independent tree-level three-dimensional reconstruction approach. XGBoost achieved the lowest errors at IFAPA (RMSE = 0.400 Mg ha−1; R2 = 0.994) and Villaseca (RMSE = 0.872 Mg ha−1; R2 = 0.995), whereas PNOA/ALS was competitive at Doña María (RMSE = 0.725 Mg ha−1; R2 = 0.994). TreeQSM closely matched the field inventory at the low-biomass IFAPA site but tended to overestimate biomass at Doña María and Villaseca, and only 28% of scanned trees yielded usable reconstructions. The results support the use of cross-platform LiDAR for orchard AGB and carbon mapping and identify the conditions under which open national LiDAR can enable scalable MRV of Mediterranean woody crops.
Abstract Understanding the allometric relationships between leaf area and other plant traits is essential for non-destructive growth monitoring and efficient crop management. However, no comprehensive study has yet modeled leaf area in quinoa ( Chenopodium quinoa Willd.) using simple morphological traits across different sowing dates. This study aimed to quantify allometric relationships between leaf area and plant height, leaf dry weight, stem dry weight, panicle dry weight, and total dry matter, and to evaluate whether these relationships are modified by sowing date. A two-year field experiment was conducted with 12 sowing dates under a randomized complete block design with three replications. Leaf area index (LAI) dynamics were described using a logistic model, and allometric relationships were fitted using power-law equations. The results showed that LAI followed a logistic trend across all sowing dates, with maximum values ranging from 2.7 to 7.9. Plant height provided the most reliable prediction of leaf area (R² = 0.83, b = 1.2), followed by leaf dry weight (R² = 0.72, b = 0.97). The allometric coefficients for stem dry weight (b = 1.54, R² = 0.74) and panicle dry weight (b = 1.95, R² = 0.71) showed greater variability. A striking finding was the exceptionally high allometric coefficient (b = 4.95) recorded on May 6 of the second year, indicating a pronounced shift in resource allocation toward leaf area expansion. Total dry matter was a weak predictor (R² = 0.54), likely due to leaf fall during the growing season. The hypothesis that sowing date modifies allometric relationships was confirmed, as evidenced by considerable variation in allometric coefficients across sowing dates. This study provides, for the first time, a comprehensive set of allometric models for quinoa across multiple sowing dates. Plant height and leaf dry weight are recommended as simple, rapid, and non-destructive indicators for leaf area estimation, facilitating improved crop monitoring and management under diverse environmental conditions.
Soil salinity is a major constraint on rice ( Oryza sativa L.) production, particularly during the yield-determining reproductive stage. Utilizing a high-throughput RGB platform, we non-destructively phenotyped a diverse panel of 294 rice accessions over two consecutive years. By extracting 60 dynamic image-based traits (i-traits) reflecting canopy architecture and stay-green capacity, and four seed-setting rate-related traits, our genome-wide association study (GWAS) identified 95 significant loci, 35.8% of which precisely co-localized with previously reported QTLs. We further prioritized OsSLT1 ( LOC_Os01g05790 ) as a candidate gene at a reproducible suggestive locus associated with leaf-rolling-related image traits. Transgenic evaluations confirmed that it acts as a positive regulator of salt tolerance at the seedling stage. OsSLT1 was mainly detected in the nucleus, and no significant changes in Na + or K + accumulation were observed in flag leaves under the tested salt-stress condition, suggesting that OsSLT1 may regulate salt tolerance through mechanisms beyond classical shoot ion accumulation. Natural variations in the OsSLT1 promoter were associated with transcriptional divergence. The Hap2 promoter haplotype showed significantly higher stress-induced transcriptional activity. Hap2 was rare in modern indica accessions, suggesting that it may represent a potentially useful genetic resource for future salt-tolerance improvement. Supplementary information The online version contains supplementary material available at 10.1007/s11032-026-01704-2.
Plant phenotyping is essential for modern crop breeding, yet traditional static image analysis fails to capture the nonlinear dynamics of plant growth. Existing time-series forecasting models exhibit notable limitations when processing multimodal data: global pooling operations may compress local 2D spatial topology of plants, and shallow feature concatenation may be insufficient for effective cross-modal semantic alignment. Moreover, current methods typically regress absolute morphological states, which may contribute to temporal lag during nonlinear growth spurts. In this paper, we propose ST-CrossGro-Former, a cross-modal residual forecasting framework for plant dynamic growth. The network removes the final global pooling and classification layers to preserve spatial topology and incorporates a scalar-guided cross-modal attention module based on the standard query-key-value formulation. This module utilizes 1D morphological features as queries to dynamically weight local visual regions, promoting multimodal feature alignment. Concurrently, a residual incremental forecasting strategy is introduced to predict short-term growth increments rather than absolute states, aiming to improve tracking sensitivity to sudden growth events. Evaluations on the UNL-CPPD maize dataset and supplementary validation on the FIP1 wheat field dataset show that the proposed model achieves competitive single-step forecasting accuracy and favorable temporal trajectory alignment compared with adapted spatiotemporal attention, graph-based, and physics-informed baselines under the evaluated settings. In particular, the FIP1 results suggest that ST-CrossGro-Former can maintain favorable height trajectory alignment under a field-acquired wheat setting, indicating its potential for helping mitigate temporal misalignment in dynamic growth forecasting.
Reproduction assets foundThe paper evaluates its ST-CrossGro-Former model on two public plant phenotyping datasets: the UNL-CPPD maize dataset (explicitly stated as publicly available with a repository URL) and the FIP1 wheat field dataset (public dataset from ETH Zürich, with its GigaScience dataset publication DOI). No author analysis code, Dataset · publicThe UNL-CPPD dataset used in this research was acquired from the UNL Plant Phenotyping Datasets repository, accessible at https://plantvision.unl.edu/datasets.Open asset ↗UNL Plant Phenotyping Datasets · UNL-CPPDlines:266-273Code / dataset availability confirmedOpenAlex · Crossref · checked 15 Sept 2026
Abstract Heat stress causes ultrastructural damage in pollen grains, leading to reduced pollen germination, pollen size and shortened pollen tube length, ultimately lowering seed set and yield. This study presents a high-throughput phenotyping framework that integrates controlled-environment pollen germination assays with deep learning–based object detection for rapid, accurate, and scalable evaluation of reproductive heat tolerance in soybean breeding programs. Sixteen soybean genotypes were grown under controlled environments at optimal (28/18°C; day/night) and high temperature (38/28°C; day/night) regimes during flowering. In vitro pollen germination was quantified using six YOLO (You Only Look Once) object-detection architectures (YOLOv7–YOLOv12) to identify the best-performing model for automated analysis. Among the tested object-detection architectures, YOLOv9 achieved the best overall performance for detecting germinated and non-germinated pollen grains in complex images. High temperature significantly reduced mean pollen germination from an average of 40% under optimal conditions to an average of 21% under heat stress (P < 0.05), with a significant genotype × growth temperature interaction. Invitro incubation temperatures ranging from 10 °C to 45 °C produced a clear thermal response; however, no significant genotype × incubation temperature interaction was detected within either growth temperature regime. Although photosynthetic and physiological traits were measured exploring their relationship with pollen germination, their transient and complex response limited their reliability for predicting reproductive performance. The automated pipeline substantially reduced the time required to evaluate pollen germination. The pipeline processed nearly 5,000 images in approximately one hour, substantially increasing throughput and reducing reliance on manual counting. The findings demonstrate that pollen germination is a promising proxy trait for screening reproductive heat tolerance in soybean. Combining controlled environment phenotyping with YOLO-based object detection enabled efficient, accurate, and scalable pollen analysis, and represents the central methodological advance of this study. YOLOv9 performed best among the tested architectures, although discrepancies from manual counts in some images indicate that additional validation is needed. The weak associations with vegetative physiological traits further support the value of direct pollen-based phenotyping.
Reproduction assets foundThe authors state that all data supporting the study, including annotated pollen germination images, computational and statistical codes, and analysis tools, were deposited in Zenodo with a public DOI. This is a paper-specific, publicly actionable asset. LabelMe and Ultralytics YOLO are generic third-party tools, not作者Dataset · publicCommission.
Data availability
All data supporting the findings of this study, including annotated images, computational and
statistical codes, and analysis tools, have been deposited in the Zenodo data repository. Additional
data will be made available upon reasonable request following acceptance of the manuscript.
Repository: https://doi.org/10.5281/zenodo.21685593
Ethics approval and consent to participate
Not applicable
Consent for publication
Not applicable
Competing Interests
Authors declared no competing interests
References
1. FAOSTAT: Crops and livestock products: soybean production data. https://www.fao.org/faostat/
(2022). Accessed 15 Feb 2026.
2. Patel D, Franklin KA. TemperaturOpen asset ↗Zenodo · 10.5281/zenodo.21685593pdf-raw-page:28 lines:1-34Code / dataset availability confirmedOpenAlex · Europe PMC · checked 15 Sept 2026
Boll-opening concentration is critical for mechanical cotton harvesting, yet it is still assessed mainly by manual records and single time-point indicators that miss temporal dynamics. To bridge the lack of a unified workflow linking vision foundation models, multi-temporal boll-opening monitoring, and harvest decision-making, we developed a cross-scale UAV high-throughput phenotyping framework centered on DINO-BollGX. DINO-BollGX couples a DINO v3 backbone, a RetinaNet detection head, and an adaptive refinement-and-suppression module for robust open-boll detection under complex field conditions. Using multi-temporal UAV imagery collected over two years for 383 cultivars, we reconstructed plot-scale time series of open-boll counts, derived dynamic features describing progression and intensity changes, and proposed a Cotton Boll-Opening Temporal Stability Index (CTSI) to quantify boll-opening rhythm and concentration; CTSI was further integrated with a time-based risk function to generate harvest decision curves. Under unified data and training settings, DINO-BollGX achieved precision = 0.91, F1 = 0.88, and AP@0.50 = 0.80, and provided accurate boll-count estimation (R 2 =0.98; MAE=3.10), outperforming YOLOv11, YOLOv12, YOLOv13, and RT-DETR. On an independent cross-year dataset acquired at 5 m altitude, it obtained precision = 0.98 and F1 = 0.87. An internal consistency analysis showed that CTSI had the expected negative association with Window_days (r = −0.83) and positive associations with Max_count (r = 0.78) and the boll-opening efficiency index (r = 0.92), reflecting the co-occurrence of temporal compactness and main-phase opening intensity in the cultivar population. CTSI ranged from −2.72 to 4.69 across cultivars, enabling identification of highly synchronized boll-opening. Harvest decision curves indicated that the relative net income index peaked at day 67 after the first observation and a compact optimal harvest window near the end of monitoring; on a fixed harvest date, Kuche 130292 (CTSI=4.69) produced 486 open bolls versus 182 for Xinluzao 36 (CTSI=0.53) and 90 for Andizhan-60 (CTSI=-2.72). Overall, the framework integrates dynamic boll-opening phenotyping with harvest timing optimization, supporting scalable cultivar screening and mechanization-ready deployment, with potential extension to harvest decision scenarios in other crops.
Reproduction assets foundThe paper publicly releases its cotton boll-opening UAV image dataset (3638 patches, 94,774 YOLO-format bounding-box annotations) on GitHub, directly supporting the paper's phenotyping analysis. No author analysis code or trained model checkpoints are explicitly deposited.Dataset · publicentary information for evaluating cross-scale detection performance and characterizing macroscopic spatial patterns. The 5 m imagery acquired on 18 Sept 2024 is used exclusively for cross-year generalization assessment. All cropped images and the corresponding YOLO-format annotation files have been publicly released on GitHub ( https://github.com/mianchen0529/cotton-boll-dataset/tree/main ) to facilitate further research on cotton phenotyping, agricultural remote sensing, and intelligent analytics.
2.3.
Model construction
2.3.1.
Overall architecture of the DINO-BollGX network
The proposed DINO-BollGX network consists of four stages: image preprocessing, feature extraction, object prediction,Open asset ↗mianchen0529/cotton-boll-datasetlines:63-74Plant phenotyping relevance matchEurope PMC · checked 15 Sept 2026
Cercospora janseana (Racib.) O. Const. is a re-emerging fungal pathogen that causes Cercospora net blotch on rice. Previous research on resistance to C. janseana has primarily focused on foliar symptoms. Subsequently, sheath infection remains poorly characterized which hinders disease management efforts. This study developed and validated a reproducible sheath inoculation protocol under controlled conditions. Three inoculation methods (agar disc, spray, and drop) were evaluated with and without mechanical wounding. Lesions only formed with inoculation methods using wounding and the agar disc method produced the most consistent and uniform symptom development. Time-course analysis in the susceptible variety Cheniere revealed earlier lesion onset, more rapid expansion, and lower variability in the agar disc method compared to spray, confirming its suitability for phenotypic screening. The optimized protocol was applied across five independent trials involving four rice varieties. DG263L consistently exhibited minimal lesion development, confirming its resistance, while Cheniere showed extensive lesion growth, indicating high susceptibility. PVL03 and LaGrue displayed moderately susceptible reactions, with PVL03 developing significantly higher lesion lengths and AUDPC values than LaGrue in one-month-old plants. Although lesion onset was delayed in 45-day-old plants, disease progressed more rapidly once established. AUDPC analysis corroborated these trends, further distinguishing varietal responses. The protocol effectively discerned resistant, intermediate, and susceptible phenotypes, supporting its use in resistance screening. To our knowledge, this is the first controlled sheath inoculation method developed for Cercospora net blotch, offering a standardized approach for evaluating sheath-specific resistance and advancing the characterization of the C. janseana -rice pathosystem.
Recent 3D foundation models provide powerful feature representations for point cloud learning by controlling spatial granularity. However, relying on a fixed spatial granularity severely limits generalization in applications like plant phenotyping, where organ morphology and size vary substantially across species and growth stages. To address this, we propose AGS-PlantSeg, a few-shot 3D plant organ segmentation method that leverages the frozen Utonia (arXiv:2603.03283) foundation model combined with Adaptive Granularity Selection. By dynamically selecting the best granularity levels for each specific plant model, our method extracts optimized geometric features for a lightweight MLP segmentation head. Extensive experiments across PLANesT-3D (arXiv:2407.21150), Pheno4D , and Crops3D demonstrate that AGS-PlantSeg significantly improves cross-species generalization, achieving 88.9% average mIoU performance and outperforming fixed-granularity baselines by 2.5 mIoU points. Despite requiring minimal annotated data, our approach is highly competitive with fully supervised, plant-specific architectures.
The edible yield and commercial value of durian are strongly determined by the number of fully developed internal locules, yet their assessment still relies largely on subjective manual inspection or costly destructive analysis. Existing non-destructive approaches, including tapping-based evaluation and X-ray imaging, are either insufficiently standardized or impractical for high-throughput phenotyping. This study proposes the Rotation-Synchronized Structural Locule Evaluator (RS-SLE). The framework integrates convolutional spatial learning with deterministic signal processing to estimate internal locule number from rotational RGB imaging. The proposed system learns frame-wise spatial morphology using convolutional heatmap regression and performs temporal inference through deterministic kinematic signal processing over a mechanically bounded 360° rotation, thereby avoiding reliance on data-intensive recurrent video models. Specifically, sequential 2D locule-probability heatmaps are transformed into a synchronized 1D morphological energy signal, which is subsequently refined using gradient-based integration, Tikhonov regularization, and adaptive morphological thresholding to recover cycle-consistent structural peaks corresponding to fertile locules. This design provides a transparent and computationally efficient alternative to learned temporal memory while maintaining robustness to viewpoint variation, partial occlusion, and high-frequency spine noise. Evaluated on 260 fruits of the ‘Monthong’ cultivar, the proposed framework achieved a mean absolute error of 0.289 locules and a 71.1% exact-match rate on independent rotational videos. When combined with morphological correlation analysis, the framework attained a 100% tolerance accuracy (±1 locule) on the test set with destructive ground-truth measurements. These results demonstrate that dynamic external morphology can serve as a reliable optical proxy for internal locule development. Crucially, the final locule count is derived entirely from the CNN-generated one-dimensional morphological signal. This demonstrates the necessity of localized structural analysis over simple macroscopic shape indices. The findings highlight the value of explainable, kinematics-informed vision systems for practical precision agriculture.
Image-based plant disease identification is essential for advancing smart agriculture, particularly for staple crops such as rice and economically significant crops such as tea, which are cultivated under complex environmental conditions. Furthermore, these crop groups present distinct challenges: rice leaf disease data typically exhibits clear pathological structures but necessitates large-scale deployment on resource-limited devices, whereas tea leaf disease data is complicated by variable lighting, diverse backgrounds, and high biodiversity. Although current deep learning models achieve high accuracy, they predominantly utilize deep convolutional neural network (CNN) architectures with millions of parameters, which hinders practical deployment on edge devices and increases the risk of overfitting when field data is scarce. In response, this study introduces DisQuan, a hybrid architecture that integrates classical deep learning with quantum machine learning (QML) to balance accuracy and resource efficiency. In particular, DisQuan combines the lightweight DisNet feature-extraction network with a variablequantum neural network to compress and refine feature representations in quantum space, yielding a model with only 0.09 million parameters. Experimental results on rice and tea leaf disease datasets indicate that DisQuan achieves the highest accuracy on the rice dataset and performance comparable to deep CNN models with significantly more parameters on the tea dataset, while maintaining a compact and stable structure. Overall, these findings suggest that DisQuan provides a practical compromise between performance and model complexity, and highlight the potential of quantum-classical hybrid architectures for plant disease detection in real-world agricultural settings and on resource-constrained devices.
In Nigeria, maize is the most widely cultivated grain, largely supporting food security for about half of the population. Research has indicated that there is annual production is declaiming to approximately 50% in maize, this is due to crop diseases and pest damage. This research presents a deep learning surveillance system for crop disease prediction and pesticides recommendations based on a DenseNet-121 architecture for continuous video streams. The research was evaluated on a curated field dataset collected from three states in Nigeria; Adamawa, Borno, and Taraba State. The system achieved a mean accuracy of 98.2% and a mean F1-score of 0.982. The results reflect a strong discriminative capacity across the diverse textural maize diseases.
Dorado-Betancourt H, van Eeuwijk F, van Etten J, Ramirez-Villegas J, de Sousa K, Daudi H, Ndegwa M, Mendes T, van Heerwaarden J.
MaizePeanut / groundnutSweet potatoField / plot
Key message Our scalable two-step method estimates genotypic performance and genetic parameters from ranking data, producing reliable results comparable to quantitative analyses, enabling the integration of ranking data into breeding pipelines. Plant breeding research has chiefly relied on on-station experiments to evaluate varietal performance. Nevertheless, these trials often fail to represent on-farm growing conditions and farmers' preferences, potentially leading to poorly defined breeding targets. Recent work has demonstrated the potential of using on-farm verification trials combined with ranking data to support farmers in evaluating varieties while providing information that is representative of farmers' needs. Despite this potential, scalable methods for quantifying genetic differences and assessing the strength of the genetic signal in such trials remain limited. Here, we present a two-step procedure for analyzing trials based on ranking data, allowing the estimation of genetic parameters. The approach follows a common strategy in quantitative genetics, in which parameters are estimated from tables of genotypic means and their variances. In our framework, these estimates are obtained from Thurstonian and/or Plackett-Luce models, which treat rankings as observations of an underlying continuous trait associated with genotypic performance. Using simulated data, we showed that genotypic mean estimates derived from ranking analyses are linearly related to those obtained from quantitative trait analyses and that their variances adequately capture estimation uncertainty. We further demonstrated that incorporating these estimates and their variances into a second-step mixed-effects model yields accurate estimates of variance components. Analyses of groundnut, maize, and sweetpotato datasets confirmed the applicability of the approach and showed that ranking data can provide reliable estimates of genetic parameters. We argue that this framework can be scaled to obtain genotypic performance estimates from multi-trial on-farm data.
Reproduction assets foundThe paper's Data availability statement provides public access to the observed groundnut and sweetpotato ranking/trial datasets (Zenodo 17112492), the authors' R functions and simulation workflow (GitHub hdorado/tricot-ranking-analysis, archived Zenodo 17942919), and supplementary material with methods and figures (ZenDataset · publicThe observed data for groundnut and sweetpotato used in this study are publicly available and can be accessed at: Global multi-crop agricultural trial data supported by citizen science, Zenodo [ https://doi.org/10.5281/zenodo.17112492 ]Open asset ↗Zenodo · 10.5281/zenodo.17112492lines:205-225Code · publicThe R functions and simulation workflow used in this study are publicly available at: - Source code available from: [ https://github.com/hdorado/tricot-ranking-analysis ]Open asset ↗GitHub · hdorado/tricot-ranking-analysislines:205-225Code · public- Archived software available from: [ https://doi.org/10.5281/zenodo.17942919 ] - License: [MIT License]Open asset ↗Zenodo · 10.5281/zenodo.17942919lines:205-225Plant phenotyping relevance matchCrossref · checked 14 Sept 2026
Published17 Aug 2026NOUN Interdisciplinary Journal of Computing, E-Learning & Application (NOUN-IJCEA)Cited by 0 · OpenAlex ↗
Manual inspection of grain plant leaves for defects is subjective and labor-intensive. Few studies have compared deep learning methods on a combined multi-crop dataset. The study collected locally 5,640 leaf images from rice, maize, and guinea corn farms in Nigeria and grouped them into six classes representing defective and healthy leaves for each crop. Three models were trained: YOLOv8 for end-to-end detection and classification, EfficientNetB0 for standalone image classification, and a hybrid that used YOLOv8 for leaf detection followed by EfficientNetB0 for patch classification. The hybrid achieved 99.85% accuracy on the test set, slightly above EfficientNetB0 (99.82%) and YOLOv8 (mAP 0.995). The hybrid also supplies bounding box locations, helping farmers identify exactly where damage appears. This system offers a reliable, field-deployable tool for monitoring grain crop health.
Rice seed vigor is a key determinant of germination performance and final crop yield, making its rapid and non-destructive assessment essential for seed quality evaluation. Conventional vigor detection methods are often destructive, labor-intensive, and time-consuming. Hyperspectral imaging provides a promising non-destructive alternative, but hyperspectral data are typically high-dimensional, redundant, and susceptible to noise and scattering interference. Moreover, existing models still have limited ability to discriminate subtle spectral differences among seed vigor levels. To address these challenges, this study proposes a gated recurrent unit (GRU)-guided closed-loop CNN-Transformer network (GCT-BCLN) for accurate, non-destructive identification of rice seed vigor. The model establishes bidirectional information flow between CNN and Transformer via the GRU, enabling dynamic and synergistic optimization of local spectral features and global spectral representations. In addition, a combined preprocessing strategy integrating adaptive iteratively reweighted penalized least squares (AirPLS), Savitzky-Golay (SG) smoothing, and multiplicative scatter correction (MSC) was adopted to improve spectral quality. Experimental results showed that GCT-BCLN achieved a test accuracy of 0.9795 for hybrid indica rice, outperforming the CNN-Transformer fusion model by 1.37%. The model also achieved accuracies of 0.9793 and 0.9758 on conventional japonica rice and glutinous japonica rice, respectively, showing consistent performance across the three evaluated variety-specific datasets under the controlled experimental protocol. These results support the feasibility of GCT-BCLN for laboratory-scale, non-destructive discrimination of aging-induced rice seed categories under controlled conditions, while practical application requires further external validation.
Red crown rot of soybean (RCR), caused by Calonectria ilicicola, is an emerging soilborne disease whose quantification is challenging due to its complex symptom development across root and foliage levels. This study developed and evaluated a multi-scale framework to improve the assessment of RCR severity from controlled environments to field conditions using root imaging and standardized visual scales. Under controlled conditions, a standard area diagram (SAD) for root necrosis was developed and validated, and SAD-assisted evaluations significantly improved accuracy, precision, and inter-rater agreement compared with unaided assessments. In field conditions, a diagrammatic symptom scale (DSS) was developed using consensus-rated images from experts and showed high reliability, repeatability, and reproducibility across 18 raters, with strong intra- and inter-rater agreement. This study developed and evaluated complementary methods to improve the assessment of RCR severity from controlled environments to field conditions using root imaging and standardized visual scales.
Quantitative flowering phenotypes are needed to support breeding and harvest management in Hypericum perforatum L. (St. John’s wort), but manual flower assessment is slow and difficult to standardize under field conditions. UAV-Hyp is a multi-temporal UAV RGB dataset containing 12,653 high-resolution images acquired at 26 measurement dates across the complete flowering period of 15 H. perforatum accessions. The images represent variable illumination, soil moisture, weed pressure, and developmental stages. The dataset provides 59,163 plant bounding boxes and 107,054 flower bounding boxes. As an application example, cascaded YOLOv8 plant and flower detectors achieved mAP@0.50:0.95 values of 0.977 and 0.950, respectively. UAV-Hyp supports scalable flower quantification and the development of time-series phenotyping methods for genotype comparison and quality-oriented medicinal-plant breeding.
Fenech M, Fernandez-Moreno J, Daubermann G, Nawar A, Taylor J, Davis H, Belcapo S, Budnick A, Yaschenko A, Xu C, Hand H, Jackson J, Vollen K, Muller K, Kater M, Moura D, Ascencio-Ibáñez J, Alonso J, Stepanova A.
ArabidopsisCell / cellular structureRootPhysiological trait estimationGrowth / development / phenology
Decoding how plants integrate multiple hormone signals to coordinate growth requires tools capable of resolving pathway interactions at cellular resolution in living tissue. Here we present ACE (Auxin–Cytokinin–Ethylene) and ACE2 , proof-of-concept single-locus reporters to simultaneously capture activity of multiple hormones. Deploying ACE alongside well-established reporters, exogenous hormone treatments, and reverse-genetic perturbations of hormone biosynthesis, signaling, and transport in three-day-old etiolated Arabidopsis seedlings, we dissect the spatiotemporal hierarchy governing primary root elongation and root apical meristem (RAM) size. We demonstrate that both ethylene- and cytokinin-triggered root growth inhibition involve a boost of TRYPTOPHAN AMINOTRANSFERASE OF ARABIDOPSIS1 (TAA1)-mediated auxin biosynthesis and AUXIN RESISTANT1 (AUX1)-dependent auxin redistribution. Two spatially distinct auxin responses underlie the respective root growth effects: ethylene expands TAA1-dependent auxin biosynthesis from the root vasculature into the epidermis and promotes AUX1-mediated auxin import into the transition and elongation zones to inhibit cell elongation, while cytokinin confines ethylene-dependent TAA1-boosted activity to the vasculature and drives auxin accumulation in lateral root cap cells to reduce RAM size. Together, these data establish a reciprocal regulatory loop between these hormones, positioning ethylene as a convergence node in auxin–cytokinin crosstalk, and cytokinin as a modulator of the ethylene–auxin interaction. Critically, the changes in cross-activated reporter patterns described for different genetic backgrounds, alongside quantitative assessment of hormone-specific inhibition of the mutants’ growth, were consistent with the multi-hormone network established over two decades of research, and added cell-type-resolved spatial detail and a proposed hierarchy for the etiolated seedling root. Finally, a second-generation reporter, ACE2 , overcomes key technical limitations of ACE , expanding the platform’s capacity toward a higher-order multi-hormone monitoring system. These resources expand the Arabidopsis genetic toolkit and provide a generalizable framework instrumental for dissecting multi-hormone signaling hierarchies at the cellular level.
Antonio Samudio Oggero · Daisy Leticia Ramírez Monzón · Héctor D. Nakayama · Luis Felipe Medeiro Alves · Valter Arthur · Oscar Vega Alvarenga · G. Resquín-Romero · Wilson Romero Vergara · Juan D. Avalos Añazco
Calibrating the mutagenic dose is the first practical step of any radiation mutation-breeding programme, and it is usually summarised by the median lethal dose (LD50) or the median growth-reduction dose (GR50). We asked whether an accessible, image-based phenotyping pipeline can quantify the early radiation response of cowpea (Vigna unguiculata L. Walp.) seedlings finely enough to estimate GR50 and to rank organ- and pigment-level sensitivities. Seeds of the traditional Paraguayan landrace kumandá pyta’i were exposed to Cobalt-60 gamma rays at 0, 100, 200, 300, 400, 500, 600, and 700 Gy, grown in a greenhouse, and photographed at the early seedling stage. A single calibrated photograph (5.1 px mm−1) of 83 seedlings was segmented in Fiji/ImageJ 1.54p and analysed with Python to extract morphometric traits (total, root, and shoot length, root:shoot ratio, tortuosity, and a two-dimensional biomass proxy) and colorimetric traits (CIE L*a*b*, a normalised greenness index, and colour-class pixel fractions). Because the data departed from normality, dose effects were tested with Kruskal–Wallis, Spearman rank correlation, and Dunn post hoc tests, and GR50 was estimated by regression of each trait expressed as a percentage of the control. Total length, shoot length, and the biomass proxy declined significantly with dose (Spearman ρ = −0.40, −0.51, and −0.47; all p < 0.001), preceded by a low-dose stimulation at 100 Gy. Estimated GR50 values were ≈390 Gy for shoot length, ≈510 Gy for total length, and ≈550 Gy for the biomass proxy, within the range reported for other cowpea genotypes. Shoot elongation was more radiosensitive than root elongation, so the root:shoot ratio did not decline; tortuosity showed no dose response. Among pigment traits, the loss of greenness was the most robust signal (a* increased, ρ = +0.62, p = 5 × 10−10; green pixel fraction fell from 0.32 to near zero by 500 Gy). These results show that single-photograph phenotyping resolves a coherent, statistically supported dose response and yields a GR50 estimate usable for dose calibration. For kumandá pyta’i, doses of roughly 300–400 Gy (below GR50) are the most defensible starting window for mutation induction. The framework is reproducible and low-cost, but it is based on one greenhouse experiment and a single genotype, and should be validated across independent trials and cultivars.
Field / plotLeafSegmentationDisease symptoms / severity
Camellia oleifera leaf disease segmentation under natural field conditions is important for precision plant protection but remains challenging because lesions often show small target areas, blurred boundaries, uneven illumination, complex backgrounds, and coexisting symptoms. To address these problems, this paper proposes BFMambaNet, a Boundary-Frequency-guided Global Semantic Mamba Network for fine-grained disease segmentation. The model adopts an encoder-decoder framework and introduces a Global Semantic Mamba-based spatial selective feature modeling block to capture long-range lesion context and reduce semantic confusion. A gated wavelet spatial enhancement block is further designed to strengthen high-frequency boundary details while suppressing noisy responses. During training, boundary-frequency auxiliary supervision guides contour localization and pathological texture recovery without additional manual boundary labels. A reinforcement-learning-guided adaptive loss controller adjusts class-wise reweighting factors and loss-component weights according to the training state, improving optimization stability. A pixel-level dataset containing 1400 images and seven disease categories was constructed for evaluation. Experimental results show that BFMambaNet achieves 92.39% Precision, 91.43% Recall, 91.26% Dice, and 85.46% mIoU, outperforming representative CNN-based, Transformer-based, and Mamba-based models. Evaluations on environmental subsets confirm superior robustness, outperforming VMamba by 3.70% mIoU under uneven illumination, 3.55% mIoU under complex backgrounds, and 5.10% mIoU under coexisting symptoms. Cross-dataset validation on Apple leaf diseases further proves its generalization with 3.39% mIoU and 3.84% Dice improvements over U-Mamba, while maintaining a competitive inference speed of 30 FPS. Qualitative results also show clearer boundaries, fewer missed small lesions, and more stable predictions in complex field scenarios.
Abstract Probe-based spatial transcriptomics platforms use predefined oligonucleotide panels to detect selected RNAs in tissue sections while preserving transcript spatial coordinates. Accurate cell segmentation is required for reliable transcript-to-cell assignments. This analytical process is affected in plant tissues by cell walls, large vacuoles, and strong autofluorescence, which often reduce boundary contrast and elevate background. Nucleus-only segmentation with fixed-distance expansion can be an alternative approach, but it underestimates cellular area and morphology and reduces the number of assignable transcripts per cell. Here, we present a practical workflow for segmentation and downstream processing in plant probe-based spatial transcriptomics. Using the soybean nodule, soybean seed, rice root, and wheat inflorescence, we demonstrate the applicability of our workflow across species, tissues, and technological platforms. In brief, candidate cell masks are generated from available fluorescence signals and then selected and corrected using two napari plugins. Transcript-informed refinement with Baysor is included as an optional step. Upon benchmarking our approach using a collection of metrics (assignment yield, background/negative controls, and per-cell transcript/gene distributions) and linked segmentation choices to expression-matrix quality and downstream clustering, we demonstrate the potential of our workflow to support the analysis of plant probe-based spatial transcriptomics.
Drought stress severely limits foxtail millet yield and quality, yet current drought-resistance indices are exclusively yield-oriented and ignore grain-filling quality. Our two-year (2024–2025) experiments with 24–48 varieties revealed that yield and blighted grain rate (BGR) are partially decoupled (e.g., Zhangzagu 18: yield 2307 kg/ha, BGR 0.444; Zhonggu 19: yield 1622 kg/ha, BGR 0.280). We therefore constructed the Yield–Quality Synergy Index (YQSI = DYI − BGR), which penalizes varieties with poor grain filling. The YQSI tied for first place with DYI in comprehensive screening performance and achieved the highest inter-annual stability (Spearman ρ = 0.823, Jaccard = 0.438, composite score = 1.261). Sensitivity analysis confirmed robustness of the equal-weight formula across a 4-fold range of quality-penalty weights. Six strongly drought-resistant germplasms with balanced yield and quality were identified. Using UAV multimodal data (RGB, multispectral, and thermal infrared) acquired during grain filling, a Random Forest model predicted a YQSI with overall R2 = 0.819 and an F1 score of 0.933 for variety screening. Feature-importance analysis highlighted NDVI, WDRVI, and red-edge texture as key predictors. This study provides a quality-constrained drought-resistance evaluation framework and demonstrates the potential of UAV-based high-throughput phenotyping for foxtail millet breeding.
Introduction Applying general-purpose vision-language models (VLMs) to crop disease diagnosis presents three critical bottlenecks: reliance on large-scale annotated data, the high computational cost of full finetuning, and existing adaptation methods designed mainly for discriminative classification without sufficient visual-linguistic interaction for generative diagnosis. Methods We propose BLAP, an adaptive multi-scale visual prompt fine-tuning framework built upon BLIP-2. BLAP introduces an adaptive visual prompt fusion module (APFM) with learnable prompt vectors and a gating mechanism, together with a multi-scale pyramid feature fusion module (PFM). All BLIP-2 backbone parameters are frozen, and only 0.11% of the model parameters are optimized. Results On a few-shot dataset comprising 990 images from 11 crops and 33 disease categories, BLAP achieved 92.78% recognition accuracy, outperforming the BLIP-2+LoRA baseline by 21.67 percentage points. BLEU-4 and ROUGE-L scores reached 0.6507 and 0.7184, respectively, while inference latency increased by only 2.15%. Discussion BLAP provides a lightweight solution that balances accuracy, efficiency, and interpretability for crop disease diagnosis in resource-constrained settings. The proposed dynamic prompt fusion and multiscale pyramid adaptation strategy may also be extended to parameter-efficient fine-tuning of visionlanguage models in other domain-specific applications.
Reproduction assets foundThe paper's few-shot crop disease dataset (990 images, 11 crops, 33 classes) is compiled entirely from four public Mendeley Data image repositories, each cited in Table 1 as the data source for specific crop/disease classes. These are the plant image inputs directly used for this paper's phenotyping/analysis. No authorDataset · publicncluding laboratory and field environments (Approximately 45% of them were captured in field environments), to enhance sample representativeness and model robustness.
Table 1
The number of collected diseases or healthy image data for each crop.
Crop
Disease
No. of images
Collection conditions
Data source
Apple
Apple scab
30
Lab
https://data.mendeley.com/datasets/tywbtsjrjv/1
Cedar apple rust
30
Lab
https://data.mendeley.com/datasets/tywbtsjrjv/1
Healthy
30
Lab
https://data.mendeley.com/datasets/tywbtsjrjv/1
Cashew
Healthy
30
Lab
https://data.mendeley.com/datasets/8fr7grr73p/1
Leaf miner
30
Lab
https://data.mendeley.com/datasets/8fr7grr73p/1
Red rust
30
Lab
https://data.mendeley.com/datasets/Open asset ↗lines:37-116Plant phenotyping relevance matchEurope PMC · checked 15 Sept 2026
Existing reviews on AI in tea production are either agriculture-generic or limited to isolated tasks. This review thoroughly compares vision technologies (RGB, hyperspectral, near-infrared, thermal, Light Detection and Ranging (LiDAR), Unmanned Aerial Vehicle (UAV)) and establishes a task-oriented algorithm selection framework for the tea industry. For small-sample or near-linear problems, traditional machine learning (ML) (support vector machine (SVM); partial least squares regression (PLSR)) remains effective. For unstructured field tasks, deep learning achieves superior performance: pest detection accuracy exceeds 97%, tea bud detection reaches 96.8% with RGB images, and hyperspectral imaging predicts nitrogen content with R 2 > 0.90 and tea polyphenols with R 2 up to 0.925. Algorithm choice further differentiates by task granularity: lightweight convolutional neural networks (CNNs) balance speed and accuracy for edge deployment at 16 fps; You Only Look Once (YOLO) series detectors enable real-time localization on mobile platforms at 93.1% accuracy, 24 ms per target. No single algorithm dominates all tea tasks; selection is a trade-off among accuracy, speed, data availability, and computational constraints. These findings outline a structured analysis of the challenges and pathways for transitioning computer vision (CV) from laboratory research toward field-deployable tools.
This study investigates the automatic classification of crop phenological stages from low-altitude UAV RGB imagery. The dataset included 11,489 images of five crops: sunflower, rapeseed, soybean, wheat, and barley. The images were annotated using the Biologische Bundesanstalt, Bundessortenamt und Chemische Industrie (BBCH) scale, with labels corresponding to either single stages or stage ranges to reflect heterogeneous field conditions and transitional crop states. A pretrained ResNet18 model was adapted to the task using transfer learning. Training was conducted in two stages: first, the classification head was optimized while the backbone remained frozen; second, the entire network was fine-tuned. The model achieved strong internal test accuracy across all crops, with 100% test accuracy for rapeseed and barley, more than 99% for the remaining crops, and a mean accuracy of 99.73% under the studied survey conditions. The results also compare favorably with previously reported studies on UAV-based phenological classification. Overall, the findings support the potential of low-altitude UAV imagery and deep learning for localized phenological assessment of selected field zones in precision agriculture, while broader deployment requires validation across independent fields, seasons, regions, and survey conditions.
Summary Understanding how plant populations respond to environmental variation through functional leaf traits remains challenging due to limitations of traditional phenotyping approaches. Hyperspectral reflectance offers a powerful high‐throughput solution, simultaneously capturing leaf biochemistry, water content, and structural properties across hundreds of wavelengths. We present a framework combining hyperspectral data, inverse modeling, and network analysis to investigate population‐level variation in Streptanthus tortuosus . Using a common garden experiment with four populations, we apply supervised methods (partial least square discriminant analysis; ridge regression) to identify which spectral features differ among populations, and an unsupervised spectral network approach to characterize how wavelength correlations are organizationally structured within each population, where we treat coordination architecture itself as a population‐level phenotype that can vary with environment. The framework detects distinct, heritable spectral signatures across populations, population differences in anthocyanins, carotenoids, Chl, water content, and population‐specific network architectures. Thermally variable environments were associated with greater spectral modularity, demonstrating that trait coordination architecture varies with climate of origin. This approach addresses the phenotyping bottleneck in evolutionary ecology, providing a scalable, high‐throughput tool for characterizing genetically based population differences in both individual traits and their coordination, with broad applications for monitoring plant population responses to climate change.
Accurate field detection of candidate tea shoots could support plantation monitoring, yield estimation, fresh-leaf assessment, and future selective-harvesting research. Wuyi rock-tea shoots are small and slender, have weak visual boundaries, and are easily confused with branches, petioles, and complex canopy backgrounds. Here, we developed YOLO11s-CSNG for candidate shoot detection in natural plantation scenes. The model combines a channel-spatial feature enhancement bottleneck, a normalized Wasserstein distance constraint for bounding-box regression, and ghost convolution layers in the detection head. We evaluated the model through detector comparisons, module ablations, and repeated training with five matched random seeds on a natural-scene dataset containing four Wuyi rock-tea cultivars. Across the five matched seeds, the mean mAP@0.5 increased from 67.37 ± 0.91% to 67.97 ± 0.90% on the validation set and from 61.29 ± 0.40% to 61.88 ± 0.66% on the internal test set. Neither paired difference was statistically significant: The 95% confidence intervals included zero, and the exact two-sided paired-permutation p values were 0.375 and 0.250, respectively. The mean mAP@0.5:0.95 did not improve. YOLO11s-CSNG retained a model size and model-only edge-inference time comparable to YOLO11s, providing a compact design for candidate shoot-region detection under the sampled field conditions.
Matuszynska, A. · Sansa, O. · Adekoya, F. J. · Akinyemi, O. O. · Anokye, E. · Bashir, O. B. · Boyny, Z. Z. F. · Chukwuka, M. K. · Corvest, E. · Dada, A. O. · DellAcqua, M. · Ehemba, G. L. · Finkbeiner, A. J. · Hamabwe, S. · Hodehou, D. A. T. · Kacheyo, O. · Kamfwa, K. · Mhango, K. J. · Abdullahi, W. M. · Munduwe, G. · Ntukidem, S. · Obisesan, O. K. · Odesina, I. S. · Ogechi, N.-U. · Olaoye, O. D. · Olayinka, M. M. · Osei-Bonsu, I. · Rilwan, K. O. · Stival, L. · Tehar, Z. · Tende, R. M. · To, J. · Ugochukwu, U. K. · Unger, A. · van Aalst, M. · Vrbic, D. · Zhang, C. · Theeuwen, T. P. J. M. · Kramer, D. M. · Kromdijk, J.
Photosynthesis is among the most consequential yet genetically complex traits in crop plants, and translating its natural variation into actionable genomic targets remains a central challenge for breeding climate-resilient varieties. To start addressing this, researchers are generating increasingly large, multi-environment field photosynthesis datasets. Yet, these data have been structurally under-analysed since their inception. Here we report the outcomes of the first dedicated hackathon focused on computational mining of such field data held in Accra, Ghana, in March 2026. Bringing together data scientists, plant physiologists, geneticists, and breeders from Europe and Africa, these interdisciplinary teams used photosynthetic data collected with hand-held fluorometers to genome-wide marker data across four crop species: cowpea (Vigna unguiculata), barley (Hordeum vulgare), common bean (Phaseolus vulgaris), and potato (Solanum tuberosum). Despite using different species and methods, independent teams identified the same three key findings. First, mechanism-informed feature engineering and dynamic modelling recover genetic signals that are not detected or discarded in standard analysis pipelines, resulting in traits with improved heritability and meaningful associations with yield. Secondly, machine learning methods proved effective at uncovering genetic associations, with temporally resolved features substantially outperforming single time-point measurements. Third, raw chlorophyll fluorescence and absorbance traces consistently contained more information and predictive power than the extracted parameters currently used. A defining feature of this event was having experimentalists and data scientists working together, enabling AI approaches to be grounded in domain knowledge and biological mechanisms rather than relying on data alone.
Continuous X-ray computed tomography (XCT) combined with digital volume correlation (DVC) is presented to quantify internal three-dimensional strain and failure dynamics in apple cortex tissue during compression, revealing how the cellular microstructure governs its mechanical response. We introduce a dimensionless number Mi that is a function of the average interfacial contact area of cells, cell wall thickness, the average cell volume and tissue porosity, to describe tissue microstructure over different development stages. Mechanical softening during maturation aligned strongly with decreasing Mi, linking microstructure to effective Young's modulus, peak stress, and toughness. DVC revealed distinctive strain-distribution signatures: in young, low-porosity tissue, strain was initially diffuse with early-onset localization indicating progressive failure, whereas mature, high-porosity tissue exhibited sharply peaked strain distributions and highly localized fracture planes indicative of brittle collapse. These findings demonstrate how pore evolution, anisotropy, and cell packing jointly determine macroscopic deformation, establishing XCT-DVC as a powerful framework for connecting plant tissue architecture to mechanical function.
Reproduction assets foundThe paper's phenotyping analysis is based on the publicly available PlantVillage plant leaf disease image dataset hosted on Kaggle (54,303 labeled leaf images across 38 classes), which the authors explicitly state was sourced from a publicly available Kaggle dataset. No author-specific code, models, or derived datasetsDataset · publicThe research incorporated PlantVillage dataset(24)
accessible on Kaggle that contains 54,303 plant leaf
images showing both healthy and diseased
conditions spanning across 38 specific categories.Open asset ↗Kagglepdf-raw-page:2 lines:1-105Plant phenotyping relevance matchEurope PMC · checked 15 Sept 2026
Published17 Aug 2026Growth hormone & IGF research : official journal of the Growth Hormone Research Society and the International IGF Research SocietyCited by 0 · OpenAlex ↗
Soliman AT, Alyafei F, Alaaraj N, Hamed N, Ahmed S.
Background ecombinant human growth hormone (rhGH) has been used for four decades under a largely population-based dosing paradigm, yet growth response varies substantially among children with apparently similar auxological phenotypes. Advances in genomics, GH-insulin-like growth factor-1 (IGF-1) axis biomarkers, mathematical and machine-learning (ML) prediction models, and digital health technologies now allow individualized characterization of growth disorders, forming the basis of an emerging precision-endocrinology paradigm. Objectives (1) to synthesize evidence on monogenic and polygenic genetic determinants of pediatric growth faltering and/or short stature and their diagnostic yield; (2) to evaluate GH-IGF-1 axis biomarkers and pharmacogenetic markers, including the IGF-1/IGFBP-3 molar ratio as a likely better indicator of bioactive IGF-1 than IGF-1 alone, together with mathematical/ML prediction models, for individualizing rhGH therapy; and (3) to appraise artificial intelligence (AI) and digital-health tools, bone-age algorithms, facial-recognition phenotyping, adherence-prediction models, and smartphone growth-monitoring, as instruments for operationalizing precision endocrinology in children and adolescents with growth disorders. Methods This is a systematic review employing narrative synthesis (a systematic scoping review). Reporting explicitly followed the PRISMA Extension for Scoping Reviews (PRISMA-ScR) checklist rather than the PRISMA 2020 statement for systematic reviews and meta-analyses, because the heterogeneous outcome metrics across genetic, diagnostic-accuracy, prediction-model, and AI/digital-health studies preclude meta-analytic pooling of a single quantitative effect size; PRISMA-ScR is the methodologically appropriate reporting framework for a review mapping evidence across such conceptually distinct domains. PubMed/MEDLINE was searched for English-language, pediatric (0-18 years) studies published between 2000 and 2026. Two-stage screening (title/abstract, then full text) was performed. Quality was appraised using design-appropriate tools: an adapted Newcastle-Ottawa Scale for genetic-association studies, QUADAS-2 for diagnostic-accuracy biomarker studies, PROBAST/TRIPOD-informed criteria for prediction-model and ML studies, and reference-standard/external-validation criteria for AI-imaging studies. Sixty-two studies were retained for qualitative synthesis. Results Monogenic defects (SHOX, ACAN, NPR2) and exome-sequencing panels explain a meaningful minority (approximately one-quarter) of previously "idiopathic" short stature, while genome-wide association studies and polygenic scores capture a substantial share of the remaining heritable variance, with polygenic risk scores achieving areas under the receiver-operating-characteristic curve up to 0.84 for predicting adult short stature. The IGF-1/IGFBP-3 M ratio outperforms IGF-1 alone for diagnosing GH deficiency (sensitivity 87.5%, specificity 83.0%), reflecting the greater bioavailability of free, unbound IGF-1 relative to that carried in the ternary IGF-1/IGFBP-3/acid-labile-subunit complex. GH-receptor exon-3 (d3) pharmacogenetic variants and machine-learning models (random forest, transcriptomic classifiers) improve prediction of individual rhGH response beyond classical mathematical models. AI-based bone-age algorithms achieve near-radiologist accuracy with reduced inter-observer variability, computer-aided facial-phenotyping tools show comparable diagnostic accuracy for syndromic short-stature disorders such as Noonan and Turner syndrome, and connected-device/ML adherence-monitoring and smartphone growth-tracking tools objectively detect suboptimal adherence and growth faltering earlier than conventional clinic-based surveillance; network meta-analyses of once-weekly long-acting rhGH formulations further suggest that reduced injection burden can translate into modestly improved height outcomes relative to daily rhGH. These findings are synthesized into a Precision-Medicine Cascade, a practice-oriented framework showing how genotype, biomarker, and digital data streams can be layered onto routine auxological assessment to guide same-visit clinical decisions on diagnostic work-up, dosing, and monitoring frequency. Conclusion Converging genetic, biomarker, computational, and digital-health evidence supports a feasible, evidence-grounded trajectory toward individualized therapy, including rhGH and emerging growth-plate-targeted agents, in pediatric growth disorders. The Precision-Medicine Cascade proposed here offers pediatric endocrinologists an immediately applicable framework for integrating these tools into everyday practice. However, current tools remain adjunctive rather than replacement for clinical judgment, and prospective, ethnically diverse validation of integrated precision-endocrinology pathways is required before routine adoption.
Hussain S, Naureen I, Habib M, Awan FS, Ghuman HF, Sher A, Riaz MW, Usman HM, Hussain S, Kong F, Zhang X.
Sustainable crop improvement is urgently needed to ensure global food security, particularly for developing and densely populated countries. The integration of artificial intelligence (AI) and machine learning (ML) into crop science tri typing is reshaping the conventional agriculture practices into an era of high-throughput phenotyping (HTPP) data-driven modern agriculture. AI tools accelerate data generation, mining, imputation, storage, transfer, and optimal decision-making within agricultural systems. AI tools are paving the way for modern plant breeding strategies by uncovering genetic variability and bridging the genotype-to-phenotype (G2P) gap, thus enabling the future of predictive breeding. Plant genetic gains or phenotype (P), by and large, depend on the genotype (G), environment (E), and their interaction (GEI). This review will provide a comprehensive overview of the historical background, current status, and prospects for integrating AI and ML tools in agricultural tri-typing, encompassing genotyping, phenotyping, and envirotyping. We explore AI-driven tools for genome analysis, HTPP platforms, and environmental data integration, emphasizing how these technologies overcome persistent bottlenecks in predictive breeding. Furthermore, this review will offer the reader key insight into modern trends, including the paradigm shift in phenomics patent filings, global distribution of HTP phenomics facilities, the publications volume and related research over the last two decades, and individual institutions currently leading or prospectively will lead the world in plant phenomics. Similar to plant phenotyping, we also try to address the integration and application of AI/ML algorithms in plant genotyping and envirotyping. Supplementary information The online version contains supplementary material available at 10.1007/s11032-026-01705-1.
Stomatal traits are key microscopic phenotypes for evaluating plant physiology, stress responses, and crop breeding potential. However, in vivo high-magnification microscopy often suffers from a shallow depth of field, causing noticeable defocus blur across different spatial locations and making it difficult to capture clear and complete stomatal structures in a single image. Multi-focus image fusion offers a practical solution, yet existing methods typically rely on supervised training, paired data, or hand-crafted rules, limiting their use in real agricultural microscopy scenarios. In this study, we propose an unsupervised multi-focus fusion framework for reconstructing fully focused stomatal microscopic images. The method integrates two-dimensional feature extraction with three-dimensional cross-focal-plane modeling to capture both spatial details and complementary information across focal planes. A max-response-guided spatial gating module is introduced to enhance focused regions while suppressing defocused responses. Additionally, dual sharpness priors based on perceptual features and wavelet high-frequency information enable pixel-wise pseudo-supervised learning without requiring all-in-focus ground-truth images. The model also predicts a probabilistic focal-plane volume for interpretable all-in-focus reconstruction. Experiments on a maize multi-focus image dataset demonstrate that the proposed method achieves superior or competitive performance across multiple fusion metrics, with entropy (EN), edge information preservation ( Q AB∕F ), Chen-Blum contrast metric ( Q CB ), and visual information fidelity for fusion (VIFF) reaching 7.43, 0.21, 0.41, and 1.01, respectively. Ablation studies confirm the effectiveness of the 3D modeling, spatial gating, and dual-prior sharpness supervision. More importantly, when the fused images serve as input to a YOLO-based stomatal instance segmentation model, the proposed method yields the best segmentation accuracy, with mAP50 and mAP50-95 reaching 0.9937 and 0.9121, respectively. Phenotypic measurements derived from the segmentation masks show high consistency with manual annotations, with the highest coefficient of determination R 2 = 0.97 achieved for stomatal count. These results indicate that the framework can act as an effective front-end module for automated microscopic stomatal phenotyping in agriculture.
Reproduction assets foundThe authors state their data and code are publicly available on GitHub, covering the multi-focus stomatal microscopy dataset and the UMF-stomata fusion/phenotyping code.Code · publicOur data and code are available at: https://github.com/Longer-S/UMF-Stomata.Open asset ↗Longer-S/UMF-Stomatahtml-lines:640-655Dataset · publicOur data and code are available at: https://github.com/Longer-S/UMF-Stomata.Open asset ↗Longer-S/UMF-Stomatahtml-lines:683-756Plant phenotyping relevance matchOpenAlex · Crossref · checked 14 Sept 2026
Photosynthesis is the fundamental biological process underlying plant growth, crop productivity, and global food security. However, its efficiency is highly vulnerable to abiotic stresses, which disrupt chlorophyll biosynthesis, electron transport, carbon assimilation, stomatal regulation, and photoprotective mechanisms, ultimately reducing crop yield. Improving photosynthetic resilience under adverse environments has therefore become a major objective of modern crop improvement. Recent advances in phenomics and high-throughput phenotyping (HTP) have transformed the evaluation of photosynthesis-related traits by enabling rapid, non-destructive, and large-scale assessment across diverse environments, while facilitating quantitative characterization of structural, physiological, biochemical, and thermal responses to abiotic stress. Technologies including chlorophyll fluorescence, gas-exchange analysis, thermal imaging, hyperspectral imaging, LiDAR, and UAV-based sensing provide comprehensive insights into plant physiological responses and stress adaptation. Integration of these phenomic approaches with genomic information and artificial intelligence (AI)-driven analytical frameworks has strengthened genomic and phenomic prediction, enabling more accurate identification of candidate genes, selection of superior genotypes, and accelerated genetic gain. This review critically synthesizes recent advances in photosynthesis-related traits, phenomics, HTP technologies, and their integration with genomics and AI-assisted breeding, highlighting current challenges, knowledge gaps, and future opportunities for developing climate-resilient wheat and rice cultivars and promoting sustainable crop production.
ABSTRACT The present study evaluated the applicability of Portable X‐ray Fluorescence (pXRF) for rapid determination of seed mineral concentrations in cowpea [ Vigna unguiculata (L.) Walp.] by comparing pXRF measurements with those obtained using Atomic Absorption Spectroscopy (AAS). Fifty‐seven cowpea genotypes, including two check varieties, were analysed for iron (Fe), zinc (Zn), manganese (Mn), copper (Cu), potassium (K), and calcium (Ca). Simple linear regression was used to assess the relationship between pXRF‐ and AAS‐derived mineral concentrations using training ( n = 47) and independent validation ( n = 10) datasets. The pXRF measurements showed good agreement with the corresponding AAS values for both macro‐ and micronutrients, with comparatively stronger relationships observed for Fe, Zn, Mn, and Cu. Residual and normal Q–Q plot analyses supported the suitability of the regression models. The findings demonstrate that pXRF enables rapid, simultaneous multielement analysis with minimal sample preparation and provides an efficient approach for high‐throughput mineral phenotyping and biofortification‐oriented cowpea breeding programmes.
Etukuri SP, Courtney CL, Rhoden C, Laufer M, Main R, Smith RA, Kenny D, Patel K, Johnson J, Thomas HB, Kothari N, Shumate B, Martin VB, Rife TW, Saski CA.
Abstract Cross-sectional analysis is considered the reference method for measuring cotton fiber fineness and maturity; however, its widespread use has been limited by labor-intensive sample preparation and manual image analysis. The objective of this study was to develop and validate a reproducible deep-learning-based workflow for automated segmentation and quantitative analysis of cotton fiber cross-sections. A total of 249 composite light microscopy images of cotton fiber cross-sections were collected and manually annotated to generate training and validation datasets. A YOLO11m instance segmentation model was developed to identify cotton fiber and lumen regions and automatically extract quantitative traits, including fiber area, lumen area, fiber perimeter, and lumen perimeter. The workflow integrates automated image segmentation, post-processing, quantitative trait extraction, and data export to facilitate reproducible cotton fiber phenotyping. Model performance was evaluated using mean Average Precision (mAP), and workflow outputs were validated against Adobe Photoshop using descriptive comparisons of six cross-sectional traits. The model achieved Box mAP50 scores of 0.984 for cotton fiber regions and 0.789 for lumen regions, demonstrating high segmentation accuracy. The automated workflow substantially reduced manual analysis time while producing measurements with central tendencies comparable to those obtained using Adobe Photoshop. To facilitate reproducibility and adoption, the workflow, trained model weights, and supporting documentation are publicly available through GitHub and a Hugging Face web application. The workflow substantially increases analytical throughput while providing a reproducible and publicly accessible method for automated cotton fiber cross-sectional phenotyping, facilitating quantitative analysis for cotton genetics and breeding research.
Reproduction assets foundThe paper's cotton fiber cross-section phenotyping workflow (YOLO11m segmentation pipeline, trained model weights, example images/outputs) is explicitly stated as publicly available via a GitHub repository and a Hugging Face web application, with URLs matching the allowed list.Code · publicThe complete source code, training scripts, dataset configuration, example input images, example outputs,
and supporting documentation are publicly available through the GitHub repository:
https://github.com/RifeLab/cotton-lumen-microOpen asset ↗RifeLab/cotton-lumen-micropdf-page:11 lines:1-47Code · publicThe cotton fiber image analysis workflow is publicly available through a web-based application hosted on
Hugging Face at: https://huggingface.co/spaces/chaneylc/cotton_fiber_microscopy_measureOpen asset ↗pdf-page:11 lines:1-47Plant phenotyping relevance matchEurope PMC · checked 15 Sept 2026
Guo Y, Ren S, Yu Q, Bai X, Ji S, Chen J, Zhu Y, Xiong J, Fang L, Liu C, Li J, Li Y, Wang Q.
Field / plotMultispectral / hyperspectralClassificationGrowth / time-series analysisGrowth / development / phenology
Aquatic plants are vital for lake ecosystem functioning and water-quality stability, yet their community dynamics and phenology rhythms remain insufficiently understood, due to the lack of effective strategies for fine-scale species mapping and phenology extraction. In this study, based on Sentinel-2 MSI imagery, we developed an integrated framework combining machine learning and a priori ecological knowledge to quantify the spatiotemporal changes in aquatic plant distribution, species composition and phenological dynamics for eight dominant species in five regulating lakes along the Eastern Route of the South-to-North Water Diversion Project in China. Results showed that the proposed framework enabled accurate aquatic plant identification, achieving an overall classification accuracy of 96.16% and over 90% accuracy for each species. Since 2016, aquatic vegetation coverage has substantially declined in most lakes, mainly due to the retreat of submerged vegetation. Community structure has shifted from submerged-plant dominance to emergent and floating-leaved dominance in two of them. Phenologically, we found that most aquatic vegetation exhibited a longer growing season, characterized by earlier growth onset (-0.28 days/year) and peak timing (-0.67 days/year) and delayed senescence (0.54 days/year). Correlation analysis indicated that aquatic vegetation dynamics was associated with climate variation, nutrient enrichment, turbidity, and water diversion, with warming and solar radiation likely promoting the growth of some emergent species, while nutrient enrichment and turbidity could be linked with submerged vegetation decline and earlier phenological shifts. Overall, this study provides an effective framework for species-level mapping and phenological monitoring of aquatic vegetation, offering valuable support for the management and conservation of lake ecosystems.
Calcium ions (Ca2+) function as ubiquitous second messengers that translate environmental and developmental cues into spatially and temporally defined cellular responses in plants. This review summarizes the cellular architecture and molecular mechanisms that generate, shape, and terminate Ca2+ signals, with emphasis on plasma-membrane channels, intracellular stores, pumps, exchangers, and organelle-associated transport systems. We also examine the development of live Ca2+ indicators, from chemical dyes and aequorin to ratiometric and single-fluorophore genetically encoded calcium indicators, and discuss principles for selecting sensors for different tissues and subcellular compartments. Recent studies have applied these tools to abiotic stress, plant immunity, polar growth, development, symbiosis, and systemic signaling. Accurate quantitative imaging nevertheless requires careful matching of sensor properties to the target cellular environment and rigorous control of motion, spectral interference, and analytical procedures. Combining improved indicators with advanced microscopy, genetic validation, and standardized data analysis should help connect distinct Ca2+ signatures with their molecular origins and physiological roles.
Efficient facility-scale tomato ripeness monitoring remains difficult in greenhouses where uneven terrain limits conventional wheeled and rail-guided platforms and planar cameras provide restricted coverage. This study developed a wheel-legged quadruped monitoring system integrating LiDAR, a depth camera, and a panoramic camera. An adaptive gait-switching strategy supported navigation across heterogeneous terrain. Panoramic images were projected into six perspective views, and the left and right views were processed using a YOLOv8-based ripeness recognition model. Time-synchronized detections and robot poses were fused to map ripeness observations into three-dimensional greenhouse coordinates. Five field experiments in a commercial tomato facility demonstrated autonomous row traversal, inter-row transition, and avoidance of pedestrians, obstacles, and cultivation boundaries. The recognition pipeline continuously identified multiple ripeness stages under variable illumination, foliage occlusion, and robot motion, while the spatial fusion procedure produced a facility-scale three-dimensional ripeness distribution. The integration of terrain-adaptive quadruped mobility, panoramic perception, and spatial mapping provides a practical framework for continuous ripeness monitoring and can support targeted harvesting, yield forecasting, and crop management.
Abstract Plants employ non-photochemical quenching (NPQ) to protect their photosynthetic apparatus from photodamage. The response latency of NPQ following changes in light intensity is thought to significantly decrease photosynthetic efficiency. The amount of NPQ is commonly quantified from chlorophyll-fluorescence techniques using the Stern–Volmer equation, which requires fully closed reaction centres (RCs) of photosystem II, yielding NPQ in the absence of photochemical quenching ( $${\rm{NPQ}}^{\rm{Closed}}$$ NPQ Closed ). However, in nature, NPQ and photochemical quenching are normally present simultaneously. Therefore, to obtain a full understanding of this process, NPQ should also be explored when the RCs are open. Here we developed two methodologies to obtain NPQ in the presence of photochemistry ( $${\rm{NPQ}}^{\rm{Open}}$$ NPQ Open ) using both fluorescence lifetime and fluorescence yield measurements. A detailed comparison in Arabidopsis thaliana plants reveals that the value of $${\mathrm{NPQ}}^{\mathrm{Open}}$$ NPQ Open is ~35% lower than that of $${\rm{NPQ}}^{\rm{Closed}}$$ NPQ Closed . This difference is consistently observed across all measurements and is seen both upon closing ( $${\rm{NPQ}}^{\rm{Open}}\to {\rm{NPQ}}^{\rm{Closed}}$$ NPQ Open → NPQ Closed ) and upon reopening ( $${\mathrm{NPQ}}^{\mathrm{Closed}}\to {\mathrm{NPQ}}^{\mathrm{Open}}$$ NPQ Closed → NPQ Open ) of the RCs. We show that this difference can be explained by the presence of RC-induced ‘instantaneous’ switching of the NPQ quenching rate. This means that, in plants, NPQ is much more economical than is widely believed, it is large when its presence is needed, and it decreases instantaneously when the need disappears.
Reproduction assets foundThe paper's custom ultrafast fluorescence analysis code (ICA-based PSII/PSI deconvolution and NPQ calculations) is explicitly deposited by the authors on GitHub, alongside the original data contributions.Code · publicr(s) for their contribution to the peer review of this work. Peer reviewer reports are available.
Funding
This work was supported by ‘Nanoscale regulators of photosynthesis’ NWO research project (project number: OCENW.GROOT.2019.86).
Data availability
The original contributions presented in the study are available via GitHub at https://github.com/L-Ramakers/Heimdall .
Code availability
The custom analysis code used in the study is available via GitHub at https://github.com/L-Ramakers/Heimdall .
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afOpen asset ↗L-Ramakers/Heimdalllines:88-125Plant phenotyping relevance matchCrossref · checked 5 Sept 2026
The uniformity of maize plant spacing serves as a critical indicator for assessing sowing quality, seed vigor, and field seedling emergence stability. However, manual measurement is inefficient, and complex field conditions make automatic seedling detection and plant spacing measurement challenging. Aiming at the challenges of missed detection, insufficient accuracy for small targets, and large errors in plant spacing calculation under complex field conditions, this study constructs a high-quality dataset containing 693 maize seedling images and implements preprocessing enhancement for images degraded by haze or dust. An intelligent maize seedling detection and plant spacing measurement method based on improved YOLOv8 is proposed. The Global Attention Mechanism (GAM) is embedded into the backbone network to strengthen cross-dimension information interaction between channels and spaces, suppress background interference, and reduce the missed detection rate. The Bi-directional Feature Pyramid Network (BiFPN) is adopted to replace the original PAFPN for enhanced multi-scale feature fusion and deep semantic representation. A new 160 × 160 high-resolution small-object detection layer is added to significantly improve the detection performance of weak and small seedlings. Experimental results demonstrate that the improved model achieves a precision, recall, mAP50, and mAP50-95 of 89.4%, 90.3%, 94.4%, and 49.4%, respectively, which are 2.6, 0.5, 1.5, and 2.7 percentage points higher than those of the original YOLOv8 model. These results indicate that the proposed model improved maize seedling detection performance under complex field conditions. Automatic plant spacing calculation is realized based on detection outputs; the average plant spacing of the dataset is 30.42 cm, with a relative error of only 4.93% compared with the preset sowing spacing of 32 cm. The proposed method can efficiently accomplish field seedling identification, plant spacing quantification, and sowing quality evaluation, providing reliable technical support for precision maize sowing, seeder parameter optimization, and intelligent field management, which is of great significance for promoting the intelligent upgrading of grain crop production.
Rodríguez-Jiménez DM, Dolinko AE, Galizzi GE, Caputo C.
BarleySeed / grainPhysiological trait estimationGrowth / development / phenologyWater status / transpiration
Background and aims Laser Biospeckle Activity (LBSA), derived from laser-induced speckle variations in response to dynamic changes in living tissues, is a promising non-invasive technique for evaluating seed germination. Methods Malting barley (Hordeum vulgare subsp. distichum L. cv. Sinfonia) seeds were analysed using five coefficients-Generalised Differences (GD), Fujii, Lasca, Frequent Motion Image (FMI), and Moment of Inertia (MI)-to assess their ability to discriminate between treatments and tissue regions, and to track changes during imbibition. Whole and longitudinally cut seeds from two treatments (control [untreated] and autoclaved [heat-inactivated]) were analysed, focusing on embryo/endosperm activity ratios. LBSA was also evaluated as a function of imbibition time and seed moisture content. Key results Four coefficients (GD, Fujii, FMI, and MI) successfully differentiated control and autoclaved seeds, as well as embryo and endosperm regions in control seeds, revealing distinct activity patterns. In control seeds, LBSA increased with imbibition time and was well described by polynomial models (quadratic for Fujii and MI; cubic for GD and FMI). GD, Fujii, and FMI required a minimum seed moisture content of 25% to detect activity, while MI was responsive only above 31.5%. In contrast, Lasca was exclusively sensitive to hydration level, fitting a relaxation curve independent of treatment. Conclusions LBSA constitutes a robust, non-destructive methodology for monitoring early germination processes. By combining coefficients, it is possible to infer physiological traits such as embryo specificity, hydration thresholds, and dynamic metabolic reactivation. This positions LBSA not only as a diagnostic tool for seed viability but also as a physiologically informative proxy for studying germination and tissue-level dynamics.
Abstract The classification of banana leaf disease has a large impact on agricultural output and relies heavily on timely early detection, with reliability as a fundamental component of effective crop management. The framework proposed in this study comprises a Hamiltonian Full Node Coverage Graph Attention Network (HFNC-GAT) and Fuzzy C-Means super pixel graph learning to explain the classification of banana leaf diseases. The HFNC-GAT allows for the representation of segmented leaf areas as the nodes in a graph. This framework also makes optimal use of an attention learning model to represent the spatial dependence of diseased leaf regions, allowing it to leverage both local and global spatial dependencies. The HFNC-GAT was demonstrated through observed experiments to achieve high performance with 96.11 and 94.19 accuracy, 0.9111 Cohen's Kappa, 0.9111 MCC, 0.9344 F2-score, and 0.9939 ROC-AUC compared to the performance of conventional CNN, GCN, and baseline GAT models.
Reproduction assets foundThe paper's Dataset Availability statement explicitly declares two public Kaggle banana leaf image datasets used for the phenotyping/classification experiments: Banana Leaf Disease Dataset V4 and BananaLSD. No author analysis code or trained model is reported as publicly available.Dataset · publicThe Banana Leaf Disease
Dataset V4 is available at https://www.kaggle.com/datasets/rayhanarlistya/banana-leaf-disease-dataset-v4.Open asset ↗Kaggle · banana-leaf-disease-dataset-v4pdf-page:19 lines:1-55Dataset · publicThe Banana Leaf Spot Diseases (BananaLSD) dataset is available at
https://www.kaggle.com/datasets/shifatearman/bananalsdOpen asset ↗Kaggle · bananalsdpdf-page:19 lines:1-55Code / dataset availability confirmedCrossref · checked 11 Sept 2026
Abstract Purpose Long-term monitoring of crop biophysical and biochemical traits remains challenging in high-latitude regions due to short growing seasons, frequent cloud cover, and highly variable weather. In this context, unmanned aerial vehicles (UAVs) offer flexible, high-resolution observations, but their added value relative to low-cost proximal sensors and their effectiveness for radiative transfer model (RTM) inversion across diverse crop canopies remain insufficiently quantified. This study evaluated the potential of a two-band proximal spectral reflectance sensor (SRS) and a five-band multispectral UAV sensor for retrieving leaf area index (LAI), leaf chlorophyll content (LCC), and canopy chlorophyll content (CCC) using PROSAIL inversion across major crops in Northern Europe over two growing seasons (2023–2024). Methods and Results Two inversion approaches – look-up table (LUT) and artificial neural network (ANN) were applied to PROSAIL simulations. UAV–PROSAIL–ANN outperformed LUT-based inversion and SRS observations, achieving the highest accuracy for LAI (R 2 = 0.81–0.95; RMSE = 0.27–0.77 m 2 /m 2 ), followed by CCC (R 2 = 0.58–0.94; RMSE 2 ), while LCC remained less accurately estimated (R 2 = 0.26–0.78; RMSE 2 ). Across sensors and methods, retrieval accuracy decreased in the order of LAI, CCC, and LCC, reflecting the stronger spectral control of canopy structure compared to biochemical traits. Conclusions The UAV–PROSAIL–ANN framework effectively captured spatial and temporal variability in crop traits, producing canopy-scale maps consistent with field observations. These results demonstrate the robustness and scalability of hybrid PROSAIL–ANN inversion for high-latitude crop monitoring, while highlighting current limitations in biochemical trait retrieval using multispectral data.
Reproduction assets foundThe paper's Data availability statement explicitly deposits the authors' UAV image processing code (irradiance normalization, vignetting, exposure compensation, radiometric calibration) in a public GitHub repository under GPL v3.0; other data are available only upon request.Code · publicData availability Code to perform irradiance normalization, vignetting, exposure compensation, and radio-
metric calibration is available at https://github.com/fieldSITES/scripts/tree/main/UAV under GNU General
Public License v3.0. Other data will be made available upon request.Open asset ↗UAVpdf-page:34 lines:1-40Plant phenotyping relevance matchCrossref · checked 11 Sept 2026
Md Nahidur Rahaman · Abdullah Al Mamun · Md. Kamal Hossen · Abdur Rouf · Tumpa Rani Shaha · Jungpil Shin · Mohd Nizam Husen · Abu Saleh Musa Miah
RiceLeafClassificationDisease symptoms / severity
Rice leaf diseases significantly affect crop health and yield potential, creating a need for accurate and timely disease diagnosis. Many existing approaches still rely on single-stream feature extraction architectures, which may limit the ability to simultaneously capture global contextual information and fine-grained disease characteristics. Moreover, limited interpretability and decision-support capability hinder their practical deployment in real-world rice farming. To address these limitations, we employed a framework consists of two parallel feature extraction streams designed to capture different characteristics of disease patterns. The first stream uses a Swin Transformer to learn global contextual information and long-range spatial relationships across the leaf image. The second stream employs ConvNeXt to extract local texture features, including lesion details, spots, and color variations. By combining these complementary representations, the proposed framework effectively integrates global semantic information with local disease-specific features for improved classification performance. The extracted features from the two streams are fused through concatenation followed by an attention-based feature refinement module, enabling adaptive weighting of discriminative features. The refined representation is then used by a fully connected classifier for disease prediction. To enhance model interpretability, Grad-CAM visualization is incorporated to highlight disease-relevant regions and provide visual explanations for the model decisions. Furthermore, an LLM-based advisory module is integrated as a post-diagnosis decision-support component to provide contextualized disease management information and suggestions based on the predicted disease category. The generated suggestions are intended to support, rather than replace, expert agronomic recommendations and should be validated by agricultural professionals before practical application. The proposed framework was evaluated on two rice leaf disease datasets, achieving accuracies of 99.55% and 97.06% on Dataset-1 and Dataset-2, respectively, which are higher than those reported in previous studies. Additionally, five-fold cross-validation on Dataset-1 achieved an average accuracy of 98.91% ± 0.47, demonstrating the stability of the proposed approach. Cross-dataset evaluation using nine common disease classes across both datasets achieved 90.80% accuracy, indicating improved generalization across different data distributions. The proposed framework provides an accurate and explainable approach for rice leaf disease diagnosis in smart agriculture applications.