基于轨道移动式激光扫描多期点云的隧道断面变形提取方法

Extraction Method of Tunnel-section Deformation Based on Orbital Mobile Laser Scanning Multi-phase Point Cloud

针对轨道移动式激光扫描技术监测地铁隧道变形的点云数据特征和隧道断面变形信息提取与分析的问题,本文提出了一种基于轨道移动式激光扫描点云的隧道断面变形提取方法和处理流程.首先采用RANSAC圆柱检测法提取地铁结构特征部位点云,利用提取的隧道结构点云进行点云的粗配准.然后设计了一种基于权重的ICP算法的精配准方法,精配准后的隧道点云通过切片点云降维处理得到二维断面数据,利用断面的k近邻点拟合曲线提取隧道断面变形值,并采用弧线投影的方法对断面变形表达.实验验证表明,提出的基于曲线拟合隧道断面变形分析的方法和过程,能够准确高效地提取到0.7 mm以上的地铁隧道断面变形信息.

Aiming at the problem of point cloud data characteristics for monitoring subway tunnel deformation with orbital mobile laser scanning technology and the extraction and analysis of tunnel section deformation information,a deformation extraction method and processing flow of tunnel section based on orbital mobile laser scanning point cloud are proposed in this paper.Firstly,the RANSAC cylinder detection method is used to extract the point cloud of the characteristic parts of the subway structure,and the extracted tunnel structure point cloud is used for rough registration of the point cloud.Afterwards,a fine registration method based on the weight-based ICP algorithm is designed,the finely registered tunnel point cloud is processed by slice point cloud dimension reduction to obtain two-dimensional cross-section data,the deformation value of the tunnel section is extracted by fitting the curve of the k-nearest neighbor points of the section,and the section deformation is expressed by the method of arc projection.The experimental verification shows that the proposed method and process of tunnel section deformation analysis based on curve fitting can accurately and efficiently extract the deformation information more than 0.7 mm of subway tunnel section.