基于RGB彩色通道的结构化道路车道线检测

Lane Detection Based on RGB Channels

车道线检测是车辆智能辅助系统的重要组成部分,为提高检测准确性,文中采用一种基于RGB颜色特征的车道线检测方法。根据车道线颜色特征设计转移函数标记图像中的车道线区域,并应用基于形态学的边缘检测算法提取车道线边缘,最终检测出车道线。文中算法原理简单,在车道线边缘识别上,具有较高的准确度,对自动车辆车道线检测有一定的意义。

Lane detection is an important part of intelligent vehicle vision navigation system. In order to im-prove the detection accuracy, this paper proposes a detection algorithm based on color feature. Transfer function isdesigned to mark the lane regions based on the color feature of the lane in the image. Then the edge detection algo-rithm based on morphology is used to detect the edge of the lane. The algorithm has the advantages of simple princi-ple and high accuracy in the lane edge recognition, and has certain significance to the automatic vehicle lane detec-tion.