Deep Learning for Early Identification of Diseases in Tomato Leaf Crops

Early disease identification is essential for prevention in tomato leaf crops and to reduce crop loss. Unfortunately, traditional manual methods are very timeconsuming and error-prone making automated solutions necessary. This study aims to use deep learning, leading the way using Convolutional Neural Networks (CNNs) for disease detection in tomato leaves. Experiments on image classification were conducted for three different models (such as ResNet50, VGG16 and an upgraded version of VGG19) Proposed VGG19 model is giving superior performance, with an accuracy of 99.5%, precision: 99.23 % and recall :99.47% F1-Score: 99.28%. At the other end of the scale, ResNet101 and ResNet50 also performed quite well but with a little less accuracy percentage while VGG16 came at last. The results reveals that the Proposed VGG19 model is highly efficient for early disease detection in tomato crops and can be a powerful tool over precision agriculture.