Drawbacks Of Dataset For Tomato Leaf Diseases Recognition

Abstract

This paper brings out the drawbacks of dataset for Tomato leaf diseases. Diseases in tomato leaf causes major loss in economy and production as well as it reduces the quality and quantity of products. As a result farmers face a lot of  problem to control and monitoring of Tomato leaf health is treated by a number of factors and out of those factors is leaf disease and because of those factors it is difficult to detect Tomato leaf diseases in the early stage. Deep learning techniques have been used to detect Tomato leaf diseases, but the existing system failed to detect leaf diseases because existing datasets are trained and tested and on few images of leaf of a particular region. Very few diseases such as Early blight, late blight are covered in the existing dataset and segmentation is missing such as in which stage leaf is healthy or not, percentage of affected area of infected leaves in the existing dataset