一种道路标线检测新方法

摘要

Lanes provides the position of the car for drivers. First the white and yellow areas in the image with the model of Maxwell color triangle are split. In the color segmentation process dual-threshold increases the accuracy of the post-processing and removes the large area or noise that not belong to the lanes. Then it determines that the image belong to the lanes or not by calculating the position and orientation of the connected component. The experiments show that this lane detection method can identify the lanes effectively and process the edges of image accurately.%道路标线为驾驶员提供车辆的位置信息.首先以麦克斯韦颜色三角为模型分割图像中的白色和黄色区域,采用双阈值分割增加后期处理精度,同时去除不属于标线的大面积区域和噪点,然后计算图像的位置与方向判定其是否为道路标线.实验表明该方法能有效检测道路标线,更好地处理图像边缘.