A Review of Deep Learning Approaches for Leaf Disease Detection in Smart Agriculture

Advancements in agricultural technology, particularly robotics and imaging, are driving smart farming and enabling precision agriculture. Autonomous systems (drones, rovers) aid farm operations, but early plant disease detection remains critical for crop health and yield. This review explores deep learning trends for leaf disease detection, emphasizing integration with imaging sensors (e.g., RGB, multispectral) to capture plant phenotypes in diverse field conditions. The study examines the robustness and performance of different deep learning architectures in challenging environments, where factors such as variable lighting, weather conditions, and overlapping symptoms can complicate accurate disease identification. Evaluating these methods on autonomous platforms highlights the potential of AI and robotics to enhance disease management and agricultural productivity.