PCL ExtractIndices提取點雲

19年8月18号更新：經過後面的學習，相對比（與基于對應點分類的對象識别等相比）這個還是比較簡單的，結合後面的分割部分程式，豁然開朗，詳見https://blog.csdn.net/suyunzzz/article/details/99694977中的二、三。

 （具體函數流程還不太懂，看完分割再回來看）

#include <iostream>
#include <pcl/ModelCoefficients.h>
#include <pcl/io/pcd_io.h>
#include <pcl/point_types.h>
#include <pcl/sample_consensus/method_types.h>
#include <pcl/sample_consensus/model_types.h>
#include <pcl/segmentation/sac_segmentation.h>
#include <pcl/filters/voxel_grid.h>
#include <pcl/filters/extract_indices.h>
#include <pcl/visualization/pcl_visualizer.h>
#include <pcl/filters/statistical_outlier_removal.h>
int
main (int argc, char** argv)
{
  pcl::PCLPointCloud2::Ptr cloud_blob (new pcl::PCLPointCloud2), cloud_filtered_blob (new pcl::PCLPointCloud2);
  pcl::PointCloud<pcl::PointXYZ>::Ptr cloud_filtered (new pcl::PointCloud<pcl::PointXYZ>), cloud_p (new pcl::PointCloud<pcl::PointXYZ>), cloud_f (new pcl::PointCloud<pcl::PointXYZ>);
  // 填入點雲資料
  pcl::PCDReader reader;
  reader.read ("table_scene_lms400.pcd", *cloud_blob);
  std::cerr << "PointCloud before filtering: " << cloud_blob->width * cloud_blob->height << " data points." << std::endl;
  // 建立濾波器對象:使用葉大小為1cm的下采樣
  pcl::VoxelGrid<pcl::PCLPointCloud2> sor;
  sor.setInputCloud (cloud_blob);
  sor.setLeafSize (0.01f, 0.01f, 0.01f);
  sor.filter (*cloud_filtered_blob);//體素濾波(下采樣)後的點雲放置到cloud_filtered_blob

  std::cerr<<cloud_filtered_blob->width*cloud_filtered_blob->height<<std::endl;

  // 轉化為模闆點雲
  pcl::fromPCLPointCloud2 (*cloud_filtered_blob, *cloud_filtered);//将下采樣後的點雲轉換為PoinCloud類型
  std::cerr << "PointCloud after filtering: " << cloud_filtered->width * cloud_filtered->height << " data points." << std::endl;
  // 将下采樣後的資料存入磁盤
  pcl::PCDWriter writer;
  writer.write<pcl::PointXYZ> ("table_scene_lms400_downsampled.pcd", *cloud_filtered, false);
  pcl::ModelCoefficients::Ptr coefficients (new pcl::ModelCoefficients ());
  pcl::PointIndices::Ptr inliers (new pcl::PointIndices ());  //建立一個PointIndices結構體指針
  // 建立分割對象
  pcl::SACSegmentation<pcl::PointXYZ> seg;
  // 可選
  seg.setOptimizeCoefficients (true); //設定對估計的模型做優化處理
  // 必選
  seg.setModelType (pcl::SACMODEL_PLANE);//設定分割模型類别
  seg.setMethodType (pcl::SAC_RANSAC);//設定使用那個随機參數估計方法
  seg.setMaxIterations (1000);//疊代次數
  seg.setDistanceThreshold (0.01);//設定是否為模型内點的距離門檻值
  // 建立濾波器對象
  pcl::ExtractIndices<pcl::PointXYZ> extract;
  int i = 0, nr_points = (int) cloud_filtered->points.size ();
  // 當還多于30%原始點雲資料時
  while (cloud_filtered->points.size () > 0.3 * nr_points)
  {
    // 從餘下的點雲中分割最大平面組成部分
    seg.setInputCloud (cloud_filtered);
    seg.segment (*inliers, *coefficients);
    if (inliers->indices.size () == 0)
    {
      std::cerr << "Could not estimate a planar model for the given dataset." << std::endl;
      break;
    }
    // 分離内層
    extract.setInputCloud (cloud_filtered);
    extract.setIndices (inliers);
    extract.setNegative (false);
    extract.filter (*cloud_p);
    std::cerr<<"cloud_filtered: "<<cloud_filtered->size()<<std::endl;//輸出提取之後剩餘的

    std::cerr<<"----------------------------------"<<std::endl;
    //儲存
    std::cerr << "PointCloud representing the planar component: " << cloud_p->width * cloud_p->height << " data points." << std::endl;
    std::stringstream ss;
    ss << "table_scene_lms400_plane_" << i << ".pcd"; //對每一次的提取都進行了檔案儲存
    writer.write<pcl::PointXYZ> (ss.str (), *cloud_p, false);
    // 建立濾波器對象
    extract.setNegative (true);//提取外層
    extract.filter (*cloud_f);//将外層的提取結果儲存到cloud_f
    cloud_filtered.swap (cloud_f);//經cloud_filtered與cloud_f交換


      i++;
  }

  std::cerr<<"cloud_filtered: "<<cloud_filtered->size()<<std::endl;

  pcl::PointCloud<pcl::PointXYZ>::Ptr cloud_seg1(new pcl::PointCloud<pcl::PointXYZ>);
  pcl::PointCloud<pcl::PointXYZ>::Ptr cloud_seg2(new pcl::PointCloud<pcl::PointXYZ>);
  pcl::PointCloud<pcl::PointXYZ>::Ptr cloud_voxel(new pcl::PointCloud<pcl::PointXYZ>);

  pcl::io::loadPCDFile("table_scene_lms400_plane_0.pcd",*cloud_seg1);
  pcl::io::loadPCDFile("table_scene_lms400_plane_1.pcd",*cloud_seg2);
  pcl::io::loadPCDFile("table_scene_lms400_downsampled.pcd",*cloud_voxel);
  /*
  //将提取結果進行統計學濾波
  pcl::StatisticalOutlierRemoval<pcl::PointXYZ> sor1;
  sor1.setInputCloud(cloud_seg2);
  sor1.setMeanK(50);
  sor1.setStddevMulThresh(1);
  sor1.filter(*cloud_f);
  std::cerr<<cloud_f->size()<<std::endl;
*/

  pcl::visualization::PCLVisualizer::Ptr viewer(new pcl::visualization::PCLVisualizer);
  viewer->initCameraParameters();

  int v1(0);
  viewer->createViewPort(0,0,0.25,1,v1);
  viewer->setBackgroundColor(0,0,255,v1);
  pcl::visualization::PointCloudColorHandlerCustom<pcl::PointXYZ> color1(cloud_voxel,244,89,233);
  viewer->addPointCloud(cloud_voxel,color1,"cloud_voxel",v1 );

  int v2(0);
  viewer->createViewPort(0.25,0,0.5,1,v2);
  viewer->setBackgroundColor(0,255,255,v2);
  pcl::visualization::PointCloudColorHandlerCustom<pcl::PointXYZ> color2(cloud_seg1,244,89,233);
  viewer->addPointCloud(cloud_seg1,color2,"cloud_seg1",v2 ); //注意id不能一樣

  int v3(0);
  viewer->createViewPort(0.5,0,0.75,1,v3);
  viewer->setBackgroundColor(34,128,0,v3);
  pcl::visualization::PointCloudColorHandlerCustom<pcl::PointXYZ> color3(cloud_seg2,244,89,233);
  viewer->addPointCloud(cloud_seg2,color3,"cloud_seg2",v3 );

  int v4(0);
  viewer->createViewPort(0.75,0,1,1,v4);
  viewer->setBackgroundColor(0,0,255,v4);
  pcl::visualization::PointCloudColorHandlerCustom<pcl::PointXYZ> color4(cloud_filtered,244,89,233);
  viewer->addPointCloud(cloud_filtered,color4,"cloud_statical",v4 );

  viewer->addCoordinateSystem();

  viewer->spin();

  return (0);
}
           

可視化：（從左到右依次為 原資料、提取一個平面、提取第二個平面、剩餘資料）