Deep Learning Methods for Disease Identification of Cotton Plants

Abstract

Cotton is a vital cash crop, contributing significantly to the global textile industry and the livelihoods of millions of farmers worldwide. However, diseases such as bacterial blight, leaf curl virus, and whitefly infestations pose a severe threat to cotton production and quality. Timely detection and accurate identification of these diseases are crucial for implementing effective control measures and ensuring crop health by exploring multiple state-of-the-art deep learning models, including CNNs and transformers. The research utilizes a diverse dataset of cotton plant images, encompassing healthy and diseased leaves, to train and fine-tune the deep learning models and Vision transformers. Additionally, we will focus on evaluating the models’ capability to detect varying intensities of whitefly infestations, which is critical for assessing disease severity and implementing appropriate control strategies. The models were cross-validated and regularized to improve the models working. This study has the potential to contribute significantly to the field of computer vision, particularly for cotton disease detection.