Defect Identification and Classification of Tomato Leaf Using Convolutional Neural Network

Abstract

Tomatoes are the most commonly grown crop globally, and they are used in almost every kitchen. India holds second place in the production of tomatoes. Due to the various kinds of diseases, the quantity and quality of tomato crop go down. Identifying the diseases in the earlier stage is very important and will help the farmers save the crop. The first initial step is pre-processing, for the Canny edge detection method is used for detecting the edges in the tomato leaves. The classification of tomato leaves is to be carried out by extracting the features like color, shape, and texture. Extracted features from segmented images are fed into classification. The convolutional neural network algorithm will be used, which will give a better accuracy to classify the diseases in the tomato leaves.