Plant disease detection using CNNs and GANs as an augmentative approach

Abstract

Almost 40% of the world's crop yield is lost to diseases and pest infestations. According to a 2012 survey, Maharashtra has the highest rate of farmer suicides and one of the major reasons for this is the failure of crops. This paper presents an image-based classification system for identification of plant diseases. Since existing datasets have diluted focus across several countries and there are none that pertain to India specifically, there is a need for establishing a local dataset to be of use to Indian farmers. It uses Generative Adversarial Networks (GANs) to augment the limited number of local images available. The classification is done by a Convolutional Neural Network (CNN) model deployed in a smart phone app.