Efficient Tomato Leaf Disease Classification Using CNN

Abstract

: Tomatoes are among the most important horticultural crops for the processing industry worldwide and play a key role in human diets due to their high nutritional and culinary value. However, plant diseases have become a major threat to global food security, increasing the need for reliable and early detection methods. Recently, research efforts have focused on developing automated systems for accurate early-stage plant disease detection to improve agricultural productivity and sustainability.With the rapid advancement of artificial intelligence (AI), particularly in machine learning and deep learning, intelligent approaches have been introduced to address complex agricultural challenges. Deep learning architectures, unlike traditional shallow neural networks, consist of multiple layers that enable more efficient feature extraction from high-dimensional image data. In this study, a deep learning–based framework is proposed for detecting and diagnosing plant diseases using images of healthy and infected leaves. The dataset consists of 22,628 images covering multiple classes of healthy and diseased plants. The proposed framework includes data preprocessing, image segmentation, data balancing, dataset splitting, classification, and performance evaluation. A fine-tuned EfficientNetB0 model is employed to enhance classification performance. Experimental results show that the proposed method achieves a mean accuracy of 99.69%, indicating strong performance and robustness. The results suggest that the model is suitable for early disease detection and has the potential for further improvement toward real-world agricultural applications. Overall, this work provides a practical foundation for developing automated plant disease detection systems.