Lane-line Detection based on Inverse Perspective Mapping and Kalman Filter

Abstract

In real-time vision-based autonomous driving, lane-lines are difficult to identify due to shadows, road disrepair, reflection or other disturbances, as well as obstacles dynamic occlusion. In this paper, region of interest (ROI) is selected first. An algorithm based on inverse perspective mapping (IPM) is proposed to transform graphic images into flat images, after the operation of gray processing and image enhancement, binarization processing and secondary image enhancement are carried out for these images. Through the pixel statistics of image columns, The characteristic curves are extracted and lane-lines are fitted. At the same time, Kalman filter algorithm is proposed to predict the lane-line which are discontinuity, this improves the robustness and stability of the lane-line detection effectively. Finally, Experiments have been conducted on different scenes, the results show that the proposed algorithm own good detection efficiency.