Tomato Leaf Disease Classification Using Deep Learning

Tomato cultivation is a vital agricultural activity worldwide, significantly contributing to food security and economic stability. However, tomato crops are highly susceptible to various diseases, leading to substantial yield losses if not detected and managed promptly. Traditional disease detection methods are often labor-intensive, time-consuming, and prone to errors, necessitating the development of automated and accurate diagnostic tools. This study proposes a deep learningbased approach for classifying tomato leaf diseases using Convolutional Neural Networks (CNNs).A dataset of 3,704 images was compiled, comprising samples of healthy leaves and leaves affected by Fusarium Wilt, Late Blight, and Tomato Yellow Leaf Curl Virus (TYLCV). The dataset included 3,200 images sourced from Kaggle and 504 images collected from field surveys in the North Wollo Zone of Ethiopia. The CNN model was trained and validated using this dataset, employing preprocessing techniques such as resizing, noise reduction, and contrast enhancement to improve image quality. The model achieved a testing accuracy of 99.08% and a loss of 0.0273 over 70 epochs, demonstrating its effectiveness in disease classification. Performance metrics, including precision, recall, and F1-score, further validated the model's robustness. The results highlight the potential of CNNs in automating tomato disease detection, offering farmers a reliable tool for early intervention and improved crop management. This research contributes to the advancement of precision agriculture, showcasing the transformative impact of deep learning technologies in sustainable farming practices.