Agriculture, vital for food security and economic stability, faces major threats from plant diseases that can reduce crop yields and affect farmer livelihoods. Early and accurate disease detection is essential for sustainable farming, yet traditional manual inspections are time-consuming, subjective, and often unavailable in rural areas. Machine learning (ML) has emerged as a transformative tool in smart agriculture, particularly for image-based plant disease recognition. Using techniques such as segmentation, feature extraction, and classification, ML algorithms-especially Convolutional Neural Networks (CNNs) and transfer learning models-can detect diseases on leaves, stems, and fruits with high accuracy. These models are increasingly deployed in real-time applications on smartphones, drones, and IoT devices. This article examines ML methodologies, case studies, and applications in plant disease detection while addressing challenges like limited datasets, environmental variability, and computational constraints, highlighting the potential of ML to improve crop management, reduce chemical misuse, and enhance food security.
M. Sabitha (Fri,) studied this question.