Plant diseases pose a significant threat to global food security by reducing crop yields and adversely affecting farmer livelihoods. Recent advancements in deep learning have transformed plant disease detection by enabling highly accurate, automated, and scalable solutions. This review systematically analyzes recent studies and classifies them into key categories, including convolutional neural network (CNN) architectures, lightweight models, hybrid and ensemble methods, transformer-based models, explainable artificial intelligence, GAN-based data augmentation, federated and edge learning, and multimodal data fusion. The paper further discusses critical challenges related to dataset diversity, model complexity, interpretability, and socioeconomic accessibility, and outlines promising future research directions. By consolidating recent developments, this review provides a comprehensive roadmap for developing scalable, interpretable, and farmer-centric plant disease detection systems.
Jeevithapriya et al. (Thu,) studied this question.