车载LiDAR点云中交通标线的提取方法

Using Vehicle-borne Laser Scanning Data for Extraction of Traffic Lines

交通标线,作为道路上重要的交通标识,为司机和行人提供重要的引导信息。车载激光扫描系统(车载LiDAR)可以快速获得被测目标的表面三维坐标信息,为提取高精度三维交通标线提供了可靠数据源。本文通过分析道路点云数据的平面距离、点云强度、点云密度等特征,将点云数据归化成地理参考强度图像。针对生成的二维参考图像,充分借鉴图像处理中目标分类与识别的手段,将交通标线信息准确提取出来。实验表明,该方法可行、有效。

Traffic lines, as important traffic signs on road, provide important information to guide drivers and pedestrians. Vehicle - borne laser scanning system can obtain three - dimensional coordinates of the target surface quickly, providing a reliable data source for extracting accuracy three - dimensional information of traffic lines. Based on the extracted road point cloud data in this article, the point cloud data is categorized into geo - referenced intensity image by analyzing the space distribution feature of the points cloud data including horizontal distance, point cloud intensity and density. Thus, the points cloud data is transformed into two - dimensional image. Then the traffic lines information can be accurately extracted by fully using the image processing methods of target classification and identification. The test result shows that the algorithm is feasible and effective.