Deep Learning Approaches for Plant Disease Management: A State-of-the-Art Analysis

Plant diseases pose a serious threat to agriculture, resulting in substantial financial losses and limiting the availability of food. Manually recognizing and classifying plant diseases may require an extensive amount of time and expertise in plant pathology. Therefore, automated disease recognition is essential for efficient disease management. This review paper analyses the state-of-the-art in the detection and classification of the diseases in the plants, with a focus on the use of computer vision, machine learning, and other techniques. This paper offers an in-depth review of several disease detection and classification approaches, including deep learning-based applications like Yolo and Conventional Neural Networks (CNN). The limitations of the systems and techniques that are now in use are also addressed by this survey. According to the review, automatic diagnosis and classification of plant diseases have shown promising outcomes through the application of computer vision and deep learning techniques. However, there are still challenges like poor dataset quantity, quality, and generalization of the classification model.