Enhanced Comparative Analysis of Tomato Leaf Disease Detection Using Machine Learning and Deep Learning Approaches

Resumen

Tomato leaf diseases significantly affect crop productivity and agricultural sustainability. Automated disease detection using artificial intelligence has emerged as a reliable solution to overcome the limitations of manual inspection. This paper presents a comparative study of tomato leaf disease detection using a traditional machine learning approach, Naive Bayes, and an advanced deep learning model, MobileNetV2. The study emphasizes both quantitative and pictorial statistical analysis to evaluate model performance. Experimental results demonstrate that MobileNetV2 achieves superior classification accuracy and robustness compared to Naive Bayes. Visual statistical representations such as accuracy comparison graphs, confusion matrices, and performance metric charts are used to clearly illustrate the effectiveness of the proposed approach.