ABSTRACT
This paper presents a novel automated approach aimed at preserving crop quality and yield by identifying illnesses in tomato leaves. The method combines Convolutional Neural Networks (CNN), Long Short-Term Memory networks (LSTM), and Single Convolutional Neural Networks (SCNN) to effectively analyze temporal patterns, focus on spatial information, and classify diseases using RGB photographs. Comparative analysis with traditional techniques demonstrates the superior effectiveness of deep learning methods. The development of a user-friendly CNN application for real-time disease prediction enhances its practical utility in agriculture, assisting farmers in managing crop health. Overall, this research significantly improves the precision and efficacy of disease detection, offering a comprehensive solution that could positively impact tomato cultivation practices.
