利用双灭点估计的车道线检测

车道线检测是自动驾驶汽车或高级驾驶辅助系统的重要组成部分，利用安装在车辆前方的单目相机以实时成像的方式获取车辆在当前车道的横向偏移，从而为车辆的车道保持、超车换道等横向控制策略提供参考。本文提出了一种基于双灭点估计的实时、稳健的车道检测方法。首先利用尺度自适应的局部对称算子对车道线点特征进行提取；其次对左右车道线分别估计灭点，以灭点为导向，构建特征点的统计直方图；然后选择穿过大多数特征点的直线且通过特征点和所选线之间的重叠度对灭点进行估计更新，重复上一步骤以获得稳定的灭点；最后基于稳定的灭点验证并选择最佳的车道线。本文在公共数据集对提出的方法进行了测试，试验结果表明，本文方法在满足实时性的前提下，能有效提高算法整体的稳健性。

Lane detection with double vanishing points estimation

In automatic vehicles or advanced driving assistance systems, lane detection provides a way to guide the vehicle to drive along its own lane. An efficient lane detection method is proposed in this paper. Firstly, the lane features are extracted, and the lines are generated from vanishing points to image bottom. Then the lines crossing most features are selected and vanishing points are updated by the overlap between features and selected lines. Last step will be repeated until stable vanishing points are obtained. The last selected lines are most likely to be lane lines. Finally, validate and select the best lane lines. The proposed method has been tested on a public dataset. The experimental results show that the method can improve robustness under real-time automated driving.