车载激光点云斑马线提取方法研究

Research on Zebra Crossing Extraction Method of Vehicle-mounted Laser Point Cloud

针对城市道路斑马线测量困难、既有基于点云数据识别提取斑马线方法存在较多弊端的问题,以车载激光扫描点云数据为基础,利用车载扫描系统点云扫描线对斑马线角点提取算法进行优化设计,通过OTSU算法分离斑马线点云,分别采用“扫描线数量-水平距离”频率分布直方图、移动判别窗口,结合RANSAC算法对斑马线长短边进行提取作业,通过斑马线长短边几何关系对斑马线角点坐标进行精确提取,依托实际工程项目对算法提取准确性进行分析验证,将斑马线角点提取结果与人工实测数据进行对比分析,通过中误差和距离均方根误差综合验证了斑马线角点提取算法的可靠性,为城市运营道路斑马线、标示线等道路特征线快速采集更新提供了新型技术方法。

In view of the difficulties in urban road zebra crossing measurement and the drawbacks of existing point cloud data identification and extraction methods,the study optimizes the zebra crossing corner extraction algorithm based on the vehicle-mounted laser scanning point cloud data,and uses the vehicle-mounted scanning system point cloud scanning line to separate the zebra crossing point cloud with OTSU algorithm.The frequency distribution histogram of“number of scanning lines-horizontal distance”and moving discrimination window are respectively used to extract the short and short edges of zebra crossings combined with RANSAC algorithm.Finally,the corner coordinates of zebra crossings are precisely extracted based on the geometric relationship between the short and long edges of zebra crossings,and the accuracy of algorithm extraction is analyzed and verified by actual engineering projects.By comparing and analyzing the results of zebra crossing corner extraction with the measured data,the reliability of the zebra crossing corner extraction algorithm is comprehensively verified through the mean error and distance root mean square error,which provides a new technical method for the rapid acquisition and updating of road feature lines,such as zebra crossing and marking lines on urban operating roads.