A Multilevel Point Cloud Classification Method for Underground Tunnels Based on Three-Dimensional Moving LiDAR Measurements

Table 1

Pseudocode of primary clustering algorithm based on unit cylinder space equation.

Step 1	The data are preliminarily filtered to remove the apparent outlier noise points.; pcl::PassThrough < pcl::PointXYZ > pass; pass.setFilterLimits (0.00, 9.22)
Step 2	The seven parameters of the unit cylinder model to be estimated are optimized.; pcl ::  SACSegmentationFromNormals < pcl ::  PointXYZ, pcl ::  Normal > seg; seg.setOptimizeCoefficients (true);
Step 3	RANSAC algorithm is used as the basic algorithm of parameter estimation.; seg.setMethodType (pcl :: SAC_RANSAC);
Step 4	Set the weight factor of the surface normal to 0.2.; seg.setNormalDistanceWeight (0.2);
Step 5	Set the maximum number of iterations to 5000.; seg.setMaxIterations (5000);
Step 6	Set the maximum allowable distance from the inner point to the model according to the estimated radius of the tunnel.; seg.setDistanceThreshold (0.1);
Step 7	According to the estimated radius setting of the tunnel, the radius range of the cylindrical model is estimated.; seg.setRadiusLimits (6.0, 7.0);