Hybrid Edge Detection Method for Autonomous Vehicle Navigation
Navjot Kaur, Kalpana Sharma, Anuj Jain · 2024
Edge detection is a fundamental job in image processing that is frequently employed in a wide range of applications. The Canny edge detection technique is wellknown for its high accuracy in detecting edges, but the Sobel edge detection approach is popular due to its quick calculation time. We present a hybrid edge detection approach in this study that incorporates the strengths of both methods. We first apply the Sobel operator to the input picture to generate an estimated edge map, which we then filter using the Canny method’s non-maximum suppression and hysteresis thresholding processes. On a set of typical benchmark photos, we next compare the performance of our hybrid technique to that of the Canny and Sobel methods. Our tests demonstrate that, when compared to the Canny approach, the suggested method not only achieves excellent accuracy in edge identification but also greatly decreases computing time.