AI-Driven Agriculture: Improving Cotton Farming Through Transfer Learning-Based Disease Detection and Classification

In the advancement of cotton farming, significant challenges related to pests and diseases have become obstacles to farming, impacting the yield and overall production of cotton. Since cotton is a cash crop, if a small percentage of the crop is affected, it can lead to a huge loss. This research paper uses various deep learning models like convolutional neural network and transfer learning models, such as the InceptionV3. Different convolutional neural network models are evaluated, and the InceptionV3, a transfer learning model, has given the best accuracy, as its complexity plays a key role in extracting features. This study used the InceptionV3 model to attain the highest validation accuracy of 96.95% in the classification of seven different diseases in cotton crops. This study is a representation of the deep learning-based capability of a disease classification system in agriculture approaches for the better accuracy of crop health and productivity. It bridges the gap between traditional farming methods and modern technology, enabling a data-driven decision-making system to ensure the sustainability of cotton farming ecosystems through healthy crops and productivity.