Open AccessTomato Leaf Disease Prediction using Convolutional Neural Network
Author(s) -
R. Sangeetha,
M. Mary Shanthi Rani
Publication year - 2019
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.l3776.119119
Subject(s) - septoria , plant disease , convolutional neural network , artificial intelligence , deep learning , machine learning , leaf spot , computer science , artificial neural network , disease , pattern recognition (psychology) , microbiology and biotechnology , biology , agronomy , medicine , pathology
Plant diseases are the common cause of the reduction in yield eventually resulting in low income to farmers. Researchers are at their best efforts to find a solution for the detection of plant diseases to increase farm productivity. In this paper, a novel approach of disease detection and prediction for tomato plant leaves has been proposed using deep learning techniques. Training of the models was performed with the use of an open database of 13,848 images, which included 7 distinct classes of [plant, disease] combinations, including healthy tomato crops. Convolution Neural Network which is well-suited for detection and prediction problems has been used for predicting healthy and unhealthy leaves affected by two types of diseases septoria spot and bacteria spot. Experiments are conducted using plant village dataset comprising of 4930 images including healthy and unhealthy leaves. The performance of the model is evaluated using precision, recall and F1-score and the model has achieved the highest accuracy 94.66 %. The significantly high success rate makes the model a very useful advisory or early warning tool.