A Deep Learning - based Approach for Tomato Plant Leaf Disease Detection and Classification

In India, agriculture is the most important industry. Agriculture is responsible for over 58 percent of rural livelihoods. Tomatoes are considered to be one among the widely grown food crops in India. As a result, disease detection on tomato plants becomes more crucial as susceptibility decreases. If sufficient care is not performed, the plant’s productivity will suffer. There are some a variety of tomato diseases that attack the plant’s leaves. When plants and harvests are harmed by bugs, it has an impact on the country’s development. Professionals usually observe the plants without the help of a microscope for the purpose of discovering and differentiating proof of illness. However, this method can be time consuming, costly, and ineffective. This paper is concerned with another method of dealing with the advancement of a tomato plant infection recognition model in light of leaf picture order, using deep convolutional networks. All of the essential improvements required to carry out this disease recognition model are fully described throughout the study, starting with gathering photographs to create a data set, which agricultural specialists evaluate, and a deep learning structure to carry out the deep CNN training. The model’s event and strangeness are its ease of use; solid leaves and foundation photos correspond to distinct classes, allowing the model to identify infected leaves as well as noise ones from the climate by utilizing CNN.