论文部分内容阅读
由于地铁盾构环片附着了大量的螺栓和螺丝以及隧道内壁上安装的大量金属支架、电器设备等附属物,使得获取的激光点云数据包含了大量的非隧道内壁点(以下简称非点),从而影响到隧道点云在形变监测、三维建模等方面的应用。本文提出基于区域分割的椭圆柱面模型方法来滤除非点,将地铁隧道横截面视为椭圆(根据盾构施工特点),利用获取的隧道原始点云数据提取出隧道中轴线,并沿隧道中轴线正交方向将点云分割为等间隔区域,然后利用各区域的点云分别迭代拟合为椭圆柱面,从而实现对隧道内壁非点的自动滤除。实验结果表明,该方法能够有效滤除隧道内的非点,为三维激光扫描技术用于地铁隧道形变监测提供高质量的点云数据。
Due to the large number of bolts and screws attached to the shield plate and the large number of metal brackets and electrical equipment attached to the inner wall of the tunnel, the acquired laser point cloud data includes a large number of non-tunnel inner wall points (hereinafter referred to as non-point) , Thus affecting the tunnel point cloud in deformation monitoring, three-dimensional modeling and other aspects of the application. In this paper, an elliptic cylindrical model based on region segmentation is proposed to filter out non-points. The cross section of the subway tunnel is regarded as an ellipse (according to the shield construction characteristics). The original tunnel point data is used to extract the central axis of the tunnel, The point cloud in the orthogonal direction of the axis is divided into equally spaced regions, and then the point cloud in each region is iteratively fitted to an elliptic cylindrical surface respectively, so as to realize the automatic filtering of the non-point inner wall of the tunnel. The experimental results show that this method can effectively filter out the non-point in the tunnel and provide high-quality point cloud data for 3D tunneling monitoring of subway tunnels.