Tomatoes are a popular food around the world, especially in Indonesia. Many tomato farmers experience crop failure due to lack of understanding and delays in recognizing diseases that attack their plants. The purpose of this study is to identify and assess the types of diseases on tomato leaves based on trends, data sources, methodologies, and characteristics used in detecting diseases on tomato leaves. The dataset used is sourced from kaggle consisting of 10 classes and contains a total of 11,000 images. The data division used consists of 90% training data and 10% test data. The augmentation and fine-tuning process is carried out to reduce over fitting. This research uses the ResNet-50 algorithm to detect and classify diseases on tomato leaves. ResNet will compare leaf images to classify them with 10 disease classes in the dataset. From the ResNet method, the average accuracy value is 93%. This shows that the ResNet-50 method for image classification can produce accurate accuracy in solving real-world problems