A Method of Multiple Lane Detection Based on Constraints of Lane Information

Multi-lane detection is one of the key technologies for advanced intelligent assisted driving. In this paper, we propose a multi-lane detection algorithm based on constraints of lane information. Unlike the methods detect multiple lanes directly, the proposed method detects ego-lane firstly and then detects the side lane, which enables the detection of multiple lanes. Specifically, the algorithm in this paper accomplishes the detection of multiple lanes in three steps. Firstly, the detection results of historical frames are used to extract the ego-lane’s region of interest(ROI), then, Hough transform and sliding window search are used for lane detection. Secondly, based on the ego-lane detection, the Hough lines of side lane are filtered by the constraints of lane information, such as the slope of lane, the range of lane width, the color of lane and the location of lane. Finally, post-processing methods are used to compensate the detection results. Experimental results in real road environment show that the algorithm is robust and has stable detection results.