Authors : Tushar Mohite Patil , Sanjay Pandey2, Ravindra Duche
DOI :10.46335/IJIES.2025.10.7.12
Abstract – In this study, we propose a machine learning-based method for automatic detection of cotton leaf diseases based on Random Forest classifier. Other features extracted are color based (RGB) and texture based (GLCM) that helps a lot in increasing the classification accuracy. It yielded a 92.5% accuracy rate, which indicates that combining these features was the right way to go! Class imbalance and similar looking diseases were tackled using data augmentation. Further work comprises applying deep learning techniques and IoT real-time based monitoring for precision agriculture. Our findings underline the power of machine learning for better diagnosis of cotton diseases and for achieving a sustainable economy.
