Automated Detection of Tomato Leaf Diseases Using Convolutional Neural Networks

In this study, an automated system to detect tomato leaf diseases with Convolutional Neural Networks (CNNs) is introduced to aid farmers in early disease identification to avert crop losses and promote yield. The dataset is multiple classes of tomato leaf diseases, and the model was trained and tuned extensively for its performance. In particular, a high training accuracy of 95.73%, and validation accuracy of 89.85% was achieved by CNN. Furthermore, the system's precision, recall, and F1 metric were around 90%, indicating the balanced system performance. The proposed CNN model is shown to be more reliable and faster than the traditional image processing techniques and hence is an appropriate choice for real time agricultural applications. The integration of modern varieties technologies in AI based approaches in agriculture which will feed productivity while maintaining Sustainability. The results indicate that the proposed system could provide an effective means for disease detection for farmers and may alleviate some of the need for chemical treatments.