Wheat leaf disease detection using CNN in Smart Agriculture

Abstract

Plant diseases are one of the most significant threats to the yield and quality of agriculture in the world. Conventional plant disease detection techniques rely heavily on expert diagnosis, which can easily lead to delays in crop disease control and field management. Therefore, early implementation of protective measures is the most effective way to control these diseases and minimize crop damage. To this end, a wheat leaf disease detection and classification method based on a Convolutional Neural Network model is proposed to increase the speed and accuracy of disease classification. The presented work used various image augmentation approaches to eliminate the overfitting problem. We achieved a testing accuracy of 94%. We also demonstrated the viability of our method for real-time detection of wheat disease in resource-constrained environments, where prompt disease detection and management are essential for sustainable agriculture.