3-D laser scanner is applied to highway survey, the point cloud density is optimized by the control of point cloud acquisition parameters, improved the accuracy of registration using ICP algorithm based on quaternion, and got rid of noise points with the algorithm of local outlier. Thus, the highway pavement point cloud of high quality is got, which can be provided as reliable data source for the follow-up design parameter extraction. Then, the road boundary lines and center lines are extracted with an edge detection algorithm and finally the design parameters are computed. Experiments show that the results fully meet the needs of the subsequent construction design.%将三维激光扫描仪应用于公路测量,通过控制点云采集的参数来优化点云密度分布,采用基于四元数的ICP算法进行配准,提高配准精度.运用局部点离群算法完成点云去噪,得到高质量的公路路面点云,为后续设计参数的提取工作提供了可靠的数据源.最后运用点云边界点识别算法准确提取了公路边界线,并进一步生成了公路中心线及各项路线设计参数.实验表明结果完全满足后续施工设计的需要.
展开▼
