点云fpfh特征分析

#include <pcl/point_types.h>
#include <pcl/features/fpfh.h>
#include <iostream>
#include <pcl/io/pcd_io.h>
#include <pcl/point_types.h>

using namespace std;
using namespace pcl;

int main()

{
    pcl::PointCloud<pcl::PointXYZ>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZ>);
    pcl::PointCloud<pcl::Normal>::Ptr normals(new pcl::PointCloud<pcl::Normal>());

    if (pcl::io::loadPCDFile<pcl::PointXYZ>("D:/point cloud coding/test/normal/monkey.pcd", *cloud) == -1) //* load the file
    {
        PCL_ERROR("Couldn't read file test_pcd.pcd \n");
        return (-1);
    }
    std::cout << "Loaded "
        << cloud->width * cloud->height
        << " data points from test_pcd.pcd with the following fields: "
        << std::endl;

    // Create the FPFH estimation class, and pass the input dataset+normals to it
    pcl::FPFHEstimation<pcl::PointXYZ, pcl::Normal, pcl::FPFHSignature33> fpfh;
    fpfh.setInputCloud(cloud);
    fpfh.setInputNormals(normals);
    // alternatively, if cloud is of tpe PointNormal, do fpfh.setInputNormals (cloud);

    // Create an empty kdtree representation, and pass it to the FPFH estimation object.
    // Its content will be filled inside the object, based on the given input dataset (as no other search surface is given).
    pcl::search::KdTree<PointXYZ>::Ptr tree(new pcl::search::KdTree<PointXYZ>);

    fpfh.setSearchMethod(tree);

    // Output datasets
    pcl::PointCloud<pcl::FPFHSignature33>::Ptr fpfhs(new pcl::PointCloud<pcl::FPFHSignature33>());

    // Use all neighbors in a sphere of radius 5cm
    // IMPORTANT: the radius used here has to be larger than the radius used to estimate the surface normals!!!
    fpfh.setRadiusSearch(0.05);

    // Compute the features
    fpfh.compute(*fpfhs);

    // fpfhs->size () should have the same size as the input cloud->size ()*
}