Performance Analysis of Tomato Leaf Disease Detection using VGG, Resnet and Mobilenet

The main goal of this study is to differentiate tomato leaves that have been harmed by certain diseases and those have not. The manual interpretation of these leaf diseases requires a lot of work and time. Many models have been created in the past to recognize various plant diseases. Deep learning and computer vision have advanced into potent agricultural technologies in recent years. A comparative analysis of tomato leaf disease detection have been conducted in this paper. The performance of the Resnet152, Resnet50, Mobilenet, MobilenetV2, VGG 19, and VGG 16 networks is compared in terms of F1_Score, Accuracy, Specificity, and Recall. The investigation concludes that the MobilenetV2 network performs better with a maximum accuracy of 97%, recall of 92%, specificity of 99%, F1 score of 95% and the miss-classification of 0.03. Therefore, this method is well suitable for tomato leaf disease detection.