Tomato leaf diseases pose a significant threat to crop yield and agricultural productivity. Addressing the challenge of automatic detection and classification of diseases found in tomato leaves through the utilization of Convolutional Neural Networks (CNNs), a subset of deep neural networks. Initially, the dataset is augmented to address the issue of limited labeled data, employing traditional data augmentation methods to balance and enlarge the dataset. Exploring the effectiveness of various fine-tuned pre-trained models, including VGG-16, VGG-19, InceptionV3, EfficientNetB7, and ResNet-50 for classifying tomato leaf diseases. Ten common diseases affecting tomato leaves are classified, with the dataset expanded fourfold through data augmentation. Experiments demonstrate that the ResNet-50 model achieves the highest accuracy of 96.15% in disease classification. Through a detailed analysis, evaluating the performance of each technique and providing insights into their effectiveness for enhancing disease classification in tomato plants.