<![CDATA[Tomato leaf diseases pose a serious threat to crop productivity and require accurate and efficient identification methods. Traditional visual inspection is time-consuming and prone to human error, motivating the need for automated image-based classification approaches. This study aims to evaluate the effectiveness of deep learning models for tomato leaf disease classification by comparing a custom Convolutional Neural Network and a transfer learning–based EfficientNet-B0 model. An experimental methodology was employed using a publicly available tomato leaf image dataset comprising nine disease classes and one healthy class. Images were preprocessed and augmented before being used to train a custom CNN and an EfficientNet-B0 model with a two-stage fine-tuning strategy. Model performance was evaluated using accuracy, precision, recall, F1-score, confusion matrices, and Receiver Operating Characteristic–Area Under Curve analysis. The experimental results demonstrate that the transfer learning approach significantly outperformed the custom CNN, with EfficientNet-B0 achieving the highest classification accuracy of 95.73% and improved class separability across disease categories.This research contributes to the field of Informatics and Computer Science by providing empirical evidence on the effectiveness of efficient transfer learning architectures for agricultural image classification. The findings support the development of resource-efficient artificial intelligence systems suitable for smart agriculture and edge-based deployment.]]>
