Date of Publication :30th May 2023

Abstract: This research paper aims to compare the performance of three popular edge detection algorithms, namely Canny, Sobel, and Laplacian, in the context of autonomous vehicles. Edge detection is a fundamental image processing task, especially in computer vision-based applications, where it is used for object detection, recognition, and tracking. Autonomous vehicles rely on computer vision algorithms to perceive the environment and make decisions accordingly. The effectiveness of edge detection algorithms directly affects the accuracy and reliability of the perception system. In this study, we evaluate the three algorithms' performance on a dataset of images captured by a camera mounted on an autonomous vehicle in various driving scenarios by comparing the effect on pictures and real-time video. We use the visual method to evaluate the algorithms' accuracy in detecting edges. Our results show that the Canny algorithm outperforms the other two algorithms in most cases, with higher precision and recall values. However, the Sobel algorithm performs better in detecting edges with lower contrast, while the Laplacian algorithm excels in detecting edges with high curvature. The findings of this study can help researchers and developers in the field of autonomous vehicles to choose the suitable edge detection algorithm based on their specific requirements

Reference :