Development and Analysis of CNN based Disease Detection in Cotton Plants

Plant diseases occur due to some organisms like bacteria, viruses and fungi, and has been a problem in agriculture around the world for centuries. Cotton is one of the most highly produced crop in India. Cotton crop help farmers to make good income. The main disadvantage of cotton crop is that it is highly prone to diseases. Early detection and diagnosis of cotton disease is a solution to this problem. Therefore, this research focuses on implementing and evaluating a Machine Learning Algorithm (CNN model) for the analysis and detection of cotton plant diseases. The dataset is pre-processed, the RGB images are converted into grayscale images and the images are resized into a fixed dimension to feed them into the CNN model. The model architecture consists of multiple convolutional layers followed by max-pooling and dense layers. The proposed method significantly contributes to the detection and management of cotton diseases, leading to increased crop yield and economic benefits for cotton farmers.