Deep Learning-Based Models For Crop Disease Detection Using Leaf Images: A Comprehensive Review

Abstract

Crop diseases pose a serious threat to agricultural productivity and global food security. Early and accurate detection of plant diseases is essential to minimize yield losses and reduce excessive pesticide usage. Traditional disease identification methods rely heavily on manual inspection by agricultural experts, which is time-consuming, subjective, and impractical for large-scale deployment. Recent advances in deep learning and computer vision have enabled automated, image-based crop disease detection with significantly improved accuracy and scalability. This review critically examines state-of-the-art deep learning techniques employed for crop disease detection using leaf images. It analyses commonly used datasets, preprocessing strategies, neural network architectures, evaluation metrics, and deployment challenges. Furthermore, existing research gaps and future directions toward robust, real-world agricultural applications are identified.

Code and data availability

This is a comprehensive review of deep learning crop disease detection; it describes public datasets (PlantVillage, PlantDoc, etc.) and cites prior works, but contains no authors' own phenotype datasets, images, code, models, or deposit statements. Cited DOIs are prior work, not paper-specific assets.

No evidence-backed public reproduction asset is currently recorded.