Automatic Identification of Tomato Disease Based on Deep Learning

Abstract

Tomato crops are vulnerable to various diseases, which can considerably impact their yield and quality. This study intends to establish an automated tomato disease recognition system utilizing deep learning methods. Convolutional Neural Networks (CNN), AlexNet, and Visual Geometry Group designs (VGG) were employed to classify tomato disease images. The designs were trained and evaluated on an extensive dataset of tomato diseases. The Visual Geometry Group model (VGG) accomplished the highest efficiency with a classification precision of 96.5%, accuracy of 95.8%, recall of 96.0%, and F1-score of 95.9%. The findings recommend that deep learning models, particularly the Visual Geometry Group model (VGG), are effective for complex image classification jobs in agricultural applications, providing valuable insights for improving automated disease detection systems in farming.