A Detection and Classification of Cotton Leaf Disease Using a Lightweight CNN Architecture

Agriculture is a major industry in many nations, including India. Because farm output accounts for a large portion of the Indian financial system, careful examination of food production issues is critical. The scientific and economic importance of crop infection nomenclature and recognition has grown in the Agricultural Industry. In the agricultural region, maintaining track of illnesses in plants with the help of experts can be very expensive. There is a need for a method or system that can automatically diagnose diseases because it has the potential to revolutionise monitoring. Massive crop fields and plant leaflets can be taken. Cotton leaf disease diagnosis is critical for preventing a catastrophic outbreak. Immediately following disease recognition, the purpose of this study is to provide guidance for the creation of an application that recognises cotton plant leaf diseases. To use this, the user must first submit a photograph of a cotton leaf, and then use image processing to obtain a digitised colour image of a damaged leaf, which may then be processed further by applying the MobileNet algorithm to anticipate the true root cause of the cotton leaf disease.

Keywords: Cotton plant, Cotton leaf, Disease, Detection, MobileNet, Feature extraction, Image classification.