一种基于半圆柱模型的隧道点云中轴线提取与滤波方法

A Method of Central Axis Extraction and Tunnel Point Cloud Filtering by a Semi-Cylindrical Model

提出一种基于半圆柱模型的隧道点云中轴线提取与滤波方法,通过粗提取和精提取中线点,采用三次B样条插值加密中线点,再采用高阶多项式精确拟合中轴线方程,根据所得中轴线方程,对点云进行切片、投影、空间圆拟合,得到隧道点云横断面,取空间圆平均半径作为隧道半径,设置距离阈值,实现隧道点云滤波。实验结果表明,该方法提取的中轴线整体偏差优于2 cm,噪声滤除率达88.969%,隧道壁点提取的正确率为91.741%,滤波效果较好。

A central axis extraction and tunnel point cloud filtering method is proposed based on a semi-cylindrical model. The midpoints are extracted through the coarse extraction and fine extraction. The cubic B-spline interpolation is used to encrypt the midpoints, and then the high-order polynomial is used to accurately fit the central axis equation. According to the obtained central axis equation, the point cloud is sliced, projected, and spatially circled to obtain the tunnel point cloud cross section. Taking the average radius of the space circle as the tunnel radius, the distance threshold is set to realize the tunnel point cloud filtering. The experimental results show that the overall accuracy of the central axis extracted by this method is better than 2 cm, the noise filtering rate is 88.969%,and the correct rate of tunnel wall point extraction is 91.741%, which verifies the filtering effect is good.