This study discusses pattern recognition using the Sobel and Canny edge detection methods. The main focus of this research is on object edge detection in images, which is important in extracting features and performing image segmentation. Sobel and Canny's edge detection method proved to be effective in pattern recognition previously. This study involved data collection using the Sobel method, training and testing of pattern recognition models, as well as experiments to determine the optimal ratio of training data and test data. The experimental results show that the Sobel edge detection method with Euclidean distance and optimal k value provides high accuracy in pattern recognition. However, it is also important to consider using the Canny edge detection method as an alternative. This research contributes to the development of pattern recognition methods that can be applied in the recognition of various and growing types of cars.