Custom Convolutional Neural Network for Cotton Disease Classification

Cotton is an organic fiber that is grown in monoculture around the world as a commercial crop. It is a vital crop on a global scale because of its economic significance, role in textiles, and effect on livelihoods. Simultaneously, cotton-related diseases have a significant impact on its productivity. The low production of these cotton crops is caused by fungal, bacterial, viral, and other parasite diseases that may change relative to the various circumstances. The loss of nutrition, environmental stress, and chemical imbalance-causing agents are the main causes of natural and common cotton diseases. To improve cotton production, it is important to detect and categorize these diseases at an early stage. So, this work creates a custom CNN model and classify the cotton diseases and compare the performance with various pre-trained model, it determines the algorithm that performs the best and categorizes cotton diseases. it is helpful to detect and classify diseases based on requirements. The findings can guide the selection of appropriate methods based on computational resources, accuracy requirements, and the need for localization information in crop disease management.