Cutting-Edge Approaches in Cotton Plant Disease Detection Using Machine Learning and Deep Learning Techniques

Agriculture is a crucial sector and income source in India. Various crops are produced for sustenance and monetary benefits. One of the most significant crops produced in many parts of India is cotton. Maharashtra is the leading cotton producer. However, numerous diseases widely affect cotton crops, especially during growth. This leads to significant yield losses. Identifying leaf disorders in the early stages is essential for the producer and the economic health of any nation involved in the cultivation. Various plant parts show symptoms when viruses, fungi, bacteria, and other pathogens infect crops. Multiple techniques, such as image processing, machine learning, and deep learning, have been researched to analyze and manage cotton pests and diseases. Convolutional Neural Networks, Visual Geometry Groups, and ResNet models work excellently in identifying leaf diseases. The performance is analyzed by F1 score, precision, and accuracy metrics. Support Vector Machine classifiers have proven reliable and efficient for this analysis, albeit with fewer images than other algorithms. This review aims to assist researchers in exploring the deep learning and machine learning techniques used for cotton plant disease analysis and beyond the classification of pests and diseases, how to work on the root cause of diseases. There is a need to treat the crop based on root cause. The root cause is a deficiency of nutrients in plants; either way, producers treat the diseases by fulfilling nutrient deficiency by applying pesticides. There is a need to treat the crop based on root cause. This paper emphasizes different techniques used in crop disease analysis.