Automated Classification of Mustard Leaf Diseases with VGG19

Mustard or brown Indian mustard is an essential crop produced worldwide for its seeds and foliage used in food or industries. Nevertheless, diseases affect the mustard plant in a way that reduces yield and quality and that is dangerous to the productivity of agriculture as well as the economy. Disease Diagnostic methods which are basic and mostly observational have poor results reaching high levels of imprecision. In pursuing this research, the VGG19 deep CNN is used in the classification of diseases ailing mustard leaves. Thus, using transfer learning, the VGG19 was further used to train the model using the entire dataset of collected mustard images which includes various forms of diseases on the leaves and the healthy ones also. The model reached good accuracy to distinguish various diseases which proved the reliability of the disease diagnosis in the shortest time. By the use of the technology that is described in this paper, farmers can be equipped with the right information for eradicating diseases affecting crops thus helping in the achievement of sustainable agriculture for the achievement of zero hunger, decent Work and economic growth, industry, innovation, and infrastructure, as well as responsible consumption and production. A loss of 0. 0143 indicates that the model has rather accurately predicted the real values that were utilized in the training portion, indicating that the model has captured all of the training data. On the training data set, the accuracy is 0.9970. On the validation set, the accuracy is 0.9308 and the loss is 0.6273.