Lane Detection Algorithm at Night Based-on Distribution Feature of Boundary Dots for Vehicle Active Safety

This study introduces a novel detection algorithm to recognize the lane markers on a structured road at night. The proposed algorithm utilizes neighborhood average filtering, Sobel operator and threshold segmentation of maximum entropy to preprocess the original image. Combining gray level image and edge image obtained by Sobel operator, we analyze the distribution feature of lane boundary dots at night and sort the boundary dots into 4 sets. Then, multiple-direction searching method is carried out to eliminate the false lane boundary dots. Final, we use adapted Hough transformation algorithm to obtain the feature parameter of the lane edge. The proposed method is proved to be reliable and robust in outside environment through experiments for the various kinds of images.