#include <iostream>
#include <pcl/io/pcd_io.h>
#include <pcl/point_types.h>
#include <pcl/visualization/cloud_viewer.h>
#include <pcl/visualization/histogram_visualizer.h>
#include <pcl/point_types.h>
#include <pcl/features/fpfh.h>
#include <pcl/filters/fast_bilateral.h>
#include <pcl/filters/passthrough.h>
#include <pcl/filters/radius_outlier_removal.h>
#include <pcl/filters/voxel_grid.h>
#include <pcl/features/normal_3d.h>
#include <pcl/visualization/pcl_plotter.h>
#include <vtkAutoInit.h>
#define vtkRenderingCore_AUTOINIT 3(vtkInteractionStyle,vtkRenderingFreeType,vtkRenderingOpenGL2)
VTK_MODULE_INIT(vtkRenderingOpenGL);
int main()
{
pcl::PointCloud<pcl::PointXYZ>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZ>);
if (pcl::io::loadPCDFile<pcl::PointXYZ>("test_pcd.pcd", *cloud) == -1) //* 读入PCD格式的文件,如果文件不存在,返回-1
{
PCL_ERROR("Couldn't read file test_pcd.pcd \n"); //文件不存在时,返回错误,终止程序。
return (-1);
}
pcl::PointCloud<pcl::PointXYZ>::Ptr cloud_fillter(new pcl::PointCloud<pcl::PointXYZ>);
pcl::PassThrough<pcl::PointXYZ> pass;
pass.setInputCloud(cloud);
pass.setFilterFieldName("z");
pass.setFilterLimits(0.0, 1.0);
pass.filter(*cloud_fillter);
pcl::PointCloud<pcl::PointXYZ>::Ptr cloud_filltered(new pcl::PointCloud<pcl::PointXYZ>);
pcl::VoxelGrid<pcl::PointXYZ> sor;
sor.setInputCloud(cloud_fillter);
sor.setLeafSize(0.01f, 0.01f, 0.01f);
sor.filter(*cloud_filltered);
/*pcl::PointCloud<pcl::PointXYZ>::Ptr cloud_outerfillter(new pcl::PointCloud<pcl::PointXYZ>);
pcl::RadiusOutlierRemoval<pcl::PointXYZ> outrem;
outrem.setInputCloud(cloud_filltered);
outrem.setRadiusSearch(0.8);
outrem.setMinNeighborsInRadius(5);
outrem.filter(*cloud_outerfillter);*/
pcl::PointCloud<pcl::Normal>::Ptr normals(new pcl::PointCloud<pcl::Normal>());
pcl::NormalEstimation<pcl::PointXYZ, pcl::Normal> ne;
ne.setInputCloud(cloud_filltered);
ne.setSearchSurface(cloud);
pcl::search::KdTree<pcl::PointXYZ>::Ptr tree(new pcl::search::KdTree<pcl::PointXYZ>());
ne.setSearchMethod(tree);
pcl::PointCloud<pcl::Normal>::Ptr cloud_normals(new pcl::PointCloud<pcl::Normal>);
ne.setRadiusSearch(0.03);
ne.compute(*cloud_normals);
pcl::FPFHEstimation<pcl::PointXYZ, pcl::Normal, pcl::FPFHSignature33> fpfh;
fpfh.setInputCloud(cloud_filltered);
fpfh.setInputNormals(cloud_normals);
pcl::search::KdTree<pcl::PointXYZ>::Ptr tree_1(new pcl::search::KdTree<pcl::PointXYZ>);
fpfh.setSearchMethod(tree_1);
pcl::PointCloud<pcl::FPFHSignature33>::Ptr fpfhs(new pcl::PointCloud<pcl::FPFHSignature33>());
fpfh.setRadiusSearch(0.05);
fpfh.compute(*fpfhs);
boost::shared_ptr<pcl::visualization::PCLVisualizer> viewer(new pcl::visualization::PCLVisualizer("3D Viewer"));
pcl::visualization::PCLPlotter plotter;
plotter.addFeatureHistogram<pcl::FPFHSignature33>(*fpfhs,"fpfh",100);
int v1(0);
viewer->createViewPort(0.0, 0.0, 0.5, 1.0, v1);
viewer->setBackgroundColor(0, 0, 0, v1);
viewer->addPointCloud<pcl::PointXYZ>(cloud, "sample cloud1", v1);
int v2(0);
viewer->createViewPort(0.5, 0.0, 1.0, 1.0, v2);
viewer->addPointCloud<pcl::PointXYZ>(cloud_filltered, "sample cloud2", v2);
viewer->setBackgroundColor(0.3, 0.3, 0.3, v2);
viewer->addPointCloudNormals<pcl::PointXYZ, pcl::Normal>(cloud_filltered, cloud_normals, 10, 0.2, "normals", v2);
viewer->initCameraParameters();
while (!viewer->wasStopped())
{
plotter.plot();
viewer->spinOnce(100);
boost::this_thread::sleep(boost::posix_time::microseconds(100000));
}
return 0;
}
关于这部分的原理可以看http://pointclouds.org/documentation/tutorials/fpfh_estimation.php#fpfh-estimation,这个PCL官网解释的还是很清楚的,其实我们只要知道这个特征的目的就好了,它是3D特征描述符的一种,所以它就类似于二维图像中的sift特征一样,可以用来做点云的识别,后期的VFH特征其实也是基于这个的。