Comparative Analysis of Algorithms for Cotton Plant Leaf Disease Classification from an Image

For disease diagnosis and yield optimization, the precise classification of pictures of cotton plant leaves into categories of healthy or diseased is a significant task in the agricultural sector. On a dataset of 2,310 photos that were made available to the public, we compared the performance of six categorization methods in this study. In terms of validation and test accuracy, our experimental results showed that ResNet50 performed better than other models, with ResNet50 reaching a validation accuracy of 0.980237 and a test accuracy of 0.9802. These results imply that ResNet is an excellent deep-learning model that can be used to accurately and dependably classify photos of cotton plant leaves in the agricultural sector. Our research demonstrates the capability of machine learning deep learning algorithms for disease diagnosis and yield improvement in the agriculture industry, which can ultimately lead to improved crop production and better food security.