Early diagnosis and treatment of tomato leaf diseases enhance plant productivity, efficiency, and quality. Misdiagnosis can lead to inadequate treatment, damaging both the plants and the agroecosystem. Therefore, accurate disease detection is essential. A rapid, precise approach to disease identification will benefit farmers significantly. Traditional manual inspection methods, while effective, are labor-intensive and prone to human error. To address these challenges, this research proposes an automated disease detection system using a custom Convolutional Neural Network (CNN). A comprehensive dataset of tomato leaves was collected, and a comparative performance analysis was conducted between YOLOv5, MobileNetV2, ResNet18, and our custom CNN model. The custom CNN model achieved an impressive accuracy of 95.2%, significantly outperforming the other models. Finally, the best-performing model was deployed in a web-based end-to-end (E2E) system, allowing tomato cultivators to classify tomato leaf diseases efficiently in real time.