TOMATO LEAF DISEASE DIAGNOSIS USING CNN

Abstract

Tomato is a horticultural crop that is cultivated globally, but it faces increasing threats from various diseases, including bacterial spot, tomato mosaic, and yellow leaf curl, which are the most prevalent and devastating, leading to significant crop losses. Early and accurate detection is crucial for effective management of these diseases. A convolutional neural network (CNN) approach has been introduced in the present study to precisely classify these specific tomato leaf diseases. A dataset comprising a total of 9,448 leaf images, representing three diseases and healthy samples, were used to train the CNN model. The evaluation metrics, including precision, recall, and F1-score values, ranged from 0.96 to 0.99, 0.97 to 1.00, and 0.98 to 1.00, respectively. The model achieved a promising accuracy of 98.99% in detecting these diseases.