Compact Convolutional Transformer (CCT)-Based Approach for Whitefly Attack Detection in Cotton Crops

Cotton is one of the world’s most economically significant agricultural products; however, it susceptible to numerous pest and virus attacks during growing season. Pests (whitefly) can significantly affect a cotton crop, but timely disease detection help control. Deep learning models are best suited for plant classification. However, data scarcity remains critical bottleneck rapidly computer vision applications. Several deep have demonstrated remarkable results in these been trained on small datasets that not reliable due model generalization issues. In this study, we first developed dataset whitefly attacked leaves containing 5135 images divided into two main classes, namely, (i) healthy (ii) unhealthy. Subsequently, proposed Compact Convolutional Transformer (CCT)-based approach classify image dataset. Experimental demonstrate CCT-based approach’s effectiveness compared state-of-the-art approaches. Our achieved an accuracy 97.2%, whereas Mobile Net, ResNet152v2, VGG-16 accuracies 95%, 92%, 90%, respectively.