存在车辆干扰的车道线识别

为避免道路上行驶的其它车辆对车道线识别的干扰,提出了一种结合车辆识别的车道线识别方法。融合雷达数据,车辆识别模块首先在图像中识别出车辆占据的区域;对于每一个车道线识别模块挑出的车道线候选点进行判断,去除处于车辆区域的车道线点;如果有效车道线点数目不足,则利用卡尔曼滤波的跟踪结果,确定符合最小风险函数的车道线位置。经过多种工况下的试验验证,该方法能够稳定地对车道线进行识别,准确地提取车道线参数,并且算法对车辆干扰有良好的抵抗能力。

Lane Detection Under Vehicles Disturbance

In order to avoid the disturbance caused by other vehicles running on the road,a lane detection method combined with vehicles detection is proposed.By fusion with radar data,the areas occupied by vehicles in the image are recognized by vehicles detection module,and the lane mark points within the vehicle occupied areas are removed from the candidate points.When the number of valid lane mark points is insufficient,additional predicted points obtained by Kalman filtering are added,and the lane mark position satisfying minimum risk function is then determined.Experiments under various conditions show that the lane detection method can stably recognize the lane marks,accurately extract the lane mark parameters,and the algorithm has rather high resistance ability to vehicles disturbance.