Journal of Advances in Biology & Biotechnology

Agriculture plays a vital role in India’s economy, with tomato cultivation often affected by various leaf diseases that can lead to significant yield losses. Timely and accurate detection is essential, yet traditional manual methods are often slow and prone to human error. This study introduces a MATLAB-based system designed to automatically detect and classify four common tomato leaf diseases Early Blight, Septoria Leaf Spot, Bacterial Wilt, and Leaf Mold using image processing techniques. The approach incorporates Otsu’s thresholding for effective image segmentation and a neural network model for disease classification, achieving an accuracy of 92% on a dataset of 14 annotated leaf images. Notably, the system achieved 88% precision in detecting Early Blight. Compared to conventional methods, this solution demonstrates improved performance and reliability. Furthermore, the system provides farmers with real-time disease identification and pesticide recommendations, offering an affordable and scalable tool that connects advanced technology with practical agricultural needs.This method has been turned into a mobile app, making it easy for farmers and users to check leaf health in the field. the app also recommends suitable pesticides tailored to the specific disease diagnosed, ensuring precise and effective treatment.