Surface Reconstruction Based on B-Spline Curve in PCL

This code run a B-spline fiting algorithm on a point cloud Q to obtain a smooth, parametric surfacere presentation.

Steps:

The B-spline surface is initialized using principal component analysis (PCA). This assumes that the poircloud has two main directions, i.e. it is roughly planar.

Refinement and fitting of B-spline surfaces

Cyclic initialization of the B-spline curve. Here, we assume that the point cloud is compact. i.e. there areno separate galaxy clusters.

Fitting of B-spline curves.

Triangulation of trimmed B-spline surfaces.

#include <iostream>
#include <pcl/point_cloud.h>
#include <pcl/point_types.h>
#include <pcl/io/pcd_io.h>
#include <vector>
#include <pcl/visualization/pcl_visualizer.h>
#include <pcl/surface/on_nurbs/fitting_surface_tdm.h>
#include <pcl/surface/on_nurbs/fitting_curve_2d_asdm.h>
#include <pcl/surface/on_nurbs/triangulation.h>
#include <pcl/console/parse.h>
using namespace pcl::console;


void PointCloud2Vector3d (pcl::PointCloud<pcl::PointXYZ>::Ptr cloud, pcl::on_nurbs::vector_vec3d &data);

void visualizeCurve (ON_NurbsCurve &curve,
                ON_NurbsSurface &surface,
                pcl::visualization::PCLVisualizer &viewer);

int main (int argc, char *argv[])
{
  
  pcl::visualization::PCLVisualizer viewer ("PCL");
  viewer.setBackgroundColor(255,255,255);

  //--------------加载点云…