A method of lane edge detection based on Canny algorithm

Lane detection is one important process in the vision-based vehicle assist system. The results of lane edge detection play an important role in feature-based lane detection. The complicated conditions of road make the correct edge detection of lane markings become very challenging. In order to get an ideal edge of lane markings in road image, a method of lane edge detection based on Canny algorithm is proposed. Firstly according to the importance in lane markings recognition, the road image is divided into three regions. Only the regions with useful information are processed. Then by the features of gray distribution and lane markings width, some noises are removed from the image. Using the shape features of lane markings, the lane edges are detected based on Canny algorithm. Finally by use of the Hough transform theory, lane detection is achieved. Experimental results show that the proposed method is effective.