Deep Learning Model for Disease Identification of Cotton Plants

Main Article Content

Rajani Zambare, Rashmi Deshmukh, Chetan Awati, Suresh Shirgave, Sandeep Thorat, Sheetal Zalte


Disease management and disease prediction from plant images can be made accurately and efficiently with the deep learning technique. Early identification of plant diseases helps farmers improve plant production and economic growth. Hence, a system that automatically detects diseases is needed to revolutionize agricultural monitoring and allow plant leaves to be healed immediately after disease identification. An experimental study shows that the proposed CNN model technique for cotton plant disease detection is efficient. The proposed system takes a cotton plant image as an input and preprocesses it to get a digitized colour image. The leaf image is segmented, and relevant features are extracted from it. The other leaf image is classified with the CNN model. Experimentation shows that the proposed CNN model achieves higher accuracy using more than three layers and three hundred epochs for training. The model is optimized by adding a dense layer and flattening. The model accuracy obtained for classification is 99.38%.

Article Details