rice 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 redistrib
Open resource ↗Zenodo · 10.5281/zenodo.21140529 · lines:868-883Paper record
IoT-RiceMobileNet: An improved lightweight MobileNetV2 Model for real-time multi-class rice disease detection using IoT.
PloS one · 15 Sept 2026 · 10.1371/journal.pone.0356383
Abstract
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.
Code and data availability
The 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.
://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.figsha
Open resource ↗GitHub · eimran/IoT-RiceMobileNet · lines:868-883the 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 resource ↗Kaggle · dedeikhsandwisaputra/rice-leafs-disease-dataset · lines:868-883ataset 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-RiceM
Open resource ↗Kaggle · anshulm257/rice-disease-dataset · lines:868-883ived 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/10.5
Open resource ↗hx6f852hw4/2 · lines:868-883