Potato Leaf Disease Detection Using CNN
Priya Khobragade, Abhishek Shriwas, Shruti Shinde, Aniruddha Mane, Ankush Padole
Abstract
During the growth season, several illnesses can affect plants. One of the most significant issues in agriculture is the early diagnosis of plant diseases. Diseases can diminish total yields and farmers' income if they are not identified early enough. Reducing plant diseases and enhancing the quality and yield of food crops can both benefit from early and accurate analysis and identification of plant diseases. Since there aren't any potato disease experts in remote places, an automated, affordable, usable, and trustworthy method is needed to diagnose plant diseases without the use of laboratory testing or expert judgement. To overcome this problem, many researchers have presented approaches based on deep learning and machine learning. However, most of these systems either use millions of training parameters or have poor classification accuracy. In this paper, we provide a model for automatic plant disease identification based on convolutional neural networks (CNN). To categories potato leaves into his three classes-healthy leaves, early blight and late blight. CNN devised a methodology. This work makes use of 5162 original image dataset and 82,592 augmented images. CNN models have a classification accuracy of 98.07% and can automatically learn features from raw photos. Convolutional neural networks (CNN) and other deep learning techniques can be used to detect plant illnesses.