基于车载LiDAR的交通标线自动提取方法研究

Research on Automatic Extraction Method of Traffic Marking Based on Vehicle LiDAR

针对自动驾驶技术对高精度道路信息实时存储分析的需求日渐增大、道路点云数据冗余离散的问题,本文提出了一种从车载LiDAR点云数据中自动提取道路面、分类并矢量化交通标线的有效方法。首先,将点云数据中的非地面点滤除;其次,基于载体车辆的行车轨迹线生成伪扫描线实现道路面的提取;然后,构建一系列二维点云参考影像,利用点云强度等特征信息检测交通标线边界像素点及坐标,并去除离群值对交通标线进行分类细化;最后,对本文方法提取与传统方法提取的交通要素进行对比,实验结果表明,本文提取方法的准确度及效率都有了一定的提升。

In view of the increasing demand of autonomous driving technology for real-time storage and analysis of high-precision road information and the redundant distization of road point cloud data,this paper proposes an effective method to automatically extract road surface,classify and vectorize traffic lines from on-board LiDAR point cloud data.Firstly,non-ground points in point cloud data are filtered out;secondly,pseudo-scanning lines are generated based on the track lines of carrier vehicles to realize road surface extraction;then,a series of 2D point cloud reference images are constructed,and the point cloud intensity and other characteristic information are used to detect the boundary pixels and coordinates of traffic markings,and the outliers are removed to classify and refine traffic markings.Finally,the traffic elements extracted by the proposed method are compared with those extracted by the traditional method.The experimental results show that the accuracy and efficiency of the proposed method have been improved.