描述

利用PCL库进行点云区域生长

代码

代码中的部分参数，还是要根据你的点云数据的实际情况，进行更改的。

• 举例子，代码中有这样两句话

  pass.setFilterFieldName ("z"); 
  pass.setFilterLimits (-1000, 1000); 
  

  按照相机的z轴方向过滤点，由于我的点单位是mm，所以是-1米到1米，如果你的点云单位是米，上面的参数很显然应该是-1和1

• 完整的main.cpp

#include <iostream>

//点云需要的头文件
#include <pcl/point_types.h>
#include <pcl/io/ply_io.h>
#include <pcl/search/search.h> 
#include <pcl/search/kdtree.h> 
#include <pcl/visualization/pcl_visualizer.h>
#include <pcl/filters/statistical_outlier_removal.h>
#include <pcl/filters/voxel_grid.h>
#include <pcl/segmentation/region_growing.h> 
#include <pcl/visualization/cloud_viewer.h> 
#include <pcl/segmentation/sac_segmentation.h>
#include <pcl/filters/passthrough.h>
#include <pcl/features/normal_3d.h>  


void drawPointCloud(pcl::PointCloud<pcl::PointXYZ>::Ptr cloud, std::string titleName)
{
    pcl::visualization::PCLVisualizer viewer (titleName);
    int v (0);

    viewer.createViewPort (0.0, 0.0, 1.0, 1.0, v);

    viewer.addCoordinateSystem(0.5);

    float bckgr_gray_level = 0.0;  // Black
    float txt_gray_lvl = 1.0 - bckgr_gray_level;

    pcl::visualization::PointCloudColorHandlerCustom<pcl::PointXYZ> cloud_in_color_h (cloud, (int) 255 * txt_gray_lvl, (int) 255 * txt_gray_lvl, (int) 255 * txt_gray_lvl);
    viewer.addPointCloud (cloud, cloud_in_color_h, "cloud_in_v1", v);

    viewer.addText (titleName, 10, 15, 16, txt_gray_lvl, txt_gray_lvl, txt_gray_lvl, "icp_info_1", v);

    viewer.setBackgroundColor (bckgr_gray_level, bckgr_gray_level, bckgr_gray_level, v);

    viewer.setCameraPosition (-3.68332, 2.94092, 5.71266, 0.289847, 0.921947, -0.256907, 0);
    viewer.setSize (1280, 1024);

    while (!viewer.wasStopped())
    {
        viewer.spinOnce();
    }
}

void regionGrowing(pcl::PointCloud<pcl::PointXYZ>::Ptr cloud_input)
{ 
    std::cout<<"[regionGrowing] input pointcloud size: "<<cloud_input->size() << std::endl; 
   
    pcl::PointCloud<pcl::PointXYZ>::Ptr cloud_downsampled(new pcl::PointCloud<pcl::PointXYZ>);
    pcl::PointCloud<pcl::PointXYZ>::Ptr cloud_processed(new pcl::PointCloud<pcl::PointXYZ>);

    //Use a voxelSampler to downsample
    pcl::VoxelGrid<pcl::PointXYZ> voxelSampler;
    voxelSampler.setInputCloud(cloud_input);
    voxelSampler.setLeafSize(0.1f, 0.1f, 0.1f);
    voxelSampler.filter(*cloud_downsampled);

    //Use a filter to reduce noise
    pcl::StatisticalOutlierRemoval<pcl::PointXYZ> statFilter;
    statFilter.setInputCloud(cloud_downsampled);
    statFilter.setMeanK(10);
    statFilter.setStddevMulThresh(0.2);
    statFilter.filter(*cloud_processed);
 
    // Create the normal estimation class, and pass the input dataset to it
    pcl::NormalEstimation<pcl::PointXYZ, pcl::Normal> ne;
    ne.setInputCloud(cloud_processed);
    // Create an empty kdtree representation, and pass it to the normal 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<pcl::PointXYZ>::Ptr tree(new pcl::search::KdTree<pcl::PointXYZ>());
    ne.setSearchMethod(tree);

    // Output datasets
    pcl::PointCloud<pcl::Normal>::Ptr normals(new pcl::PointCloud<pcl::Normal>);
    // setRadiusSearch and setKSearch are two methods of searching, both are useful, we just use setKSearch
    // ne.setRadiusSearch(0.01); // Use all neighbors in a sphere of radius 1cm
    ne.setKSearch(10); 
    // Compute the features
    ne.compute(*normals);

    pcl::IndicesPtr indices (new std::vector <int>); 
    pcl::PassThrough<pcl::PointXYZ> pass; 
    pass.setInputCloud (cloud_processed); 
    pass.setFilterFieldName ("z"); 
    pass.setFilterLimits (-1000, 1000); 
    pass.filter (*indices); 


    //聚类对象
    pcl::RegionGrowing<pcl::PointXYZ, pcl::Normal> reg; 
    reg.setMinClusterSize (5000); //最小聚类的点数 50
    reg.setMaxClusterSize (1000000);   //最大聚类的点数 1000000
    reg.setSearchMethod (tree);  //搜索方式
    reg.setNumberOfNeighbours (30); //设置搜索的邻域点的个数 30
    reg.setInputCloud (cloud_processed); //输入点云
    //reg.setIndices (indices); 
    reg.setInputNormals (normals); //输入的法线
    reg.setSmoothnessThreshold (50.0 / 180.0 * M_PI); //设置平滑度  3 
    reg.setCurvatureThreshold (1.0); //设置曲率的阈值
    std::vector <pcl::PointIndices> clusters; 
    reg.extract (clusters); 
    
    
    //输出点云簇的个数
    std::cout << "Number of clusters is equal to " << clusters.size () << std::endl; 
    std::cout << "First cluster has " << clusters[0].indices.size () << " points." << endl;
    //输出每个点云簇的点数 
    for (int i = 0 ; i<clusters.size() ; i++){
        std::cout << "ID = "<<i << " cluster has " << clusters[i].indices.size() << " points." << endl; 
    }

    
    //输出第一个点云簇的前10个点的序号
    int counter = 0;
    std::cout<<"ID = 0 cluster first 10 points id are : ";
    while (counter < clusters[0].indices.size ()) 
    { 
        std::cout <<clusters[0].indices[counter] << ", ";
        counter++; 
        if (counter == 10) 
        {
            break;
        }
    } 
    std::cout << std::endl; 


    // 用不同颜色划分各个点云簇
    pcl::PointCloud <pcl::PointXYZRGB>::Ptr colored_cloud = reg.getColoredCloud (); 
    pcl::visualization::CloudViewer viewer("Cluster viewer");
    viewer.showCloud(colored_cloud);
    while (!viewer.wasStopped())
    {
        boost::this_thread::sleep(boost::posix_time::microseconds(100000));
    }
    

    // 把想要的某一点云簇画出来
    int need = 1;
    pcl::PointCloud<pcl::PointXYZ>::Ptr cloud_objectonly(new pcl::PointCloud<pcl::PointXYZ>);
    pcl::copyPointCloud(*cloud_processed, clusters[need].indices, *cloud_objectonly);
    
    pcl::visualization::PCLVisualizer viewer_part ("Final1 with Visualization");
    int v1 (0);
    int v2 (1);
    viewer_part.createViewPort (0.0, 0.0, 0.5, 1.0, v1);
    viewer_part.createViewPort (0.5, 0.0, 1.0, 1.0, v2);

    viewer_part.addCoordinateSystem(0.5);

    float bckgr_gray_level = 0.0;  // Black
    float txt_gray_lvl = 1.0 - bckgr_gray_level;

    pcl::visualization::PointCloudColorHandlerCustom<pcl::PointXYZ> cloud_in_color_h (cloud_processed, (int) 255 * txt_gray_lvl, (int) 255 * txt_gray_lvl, (int) 255 * txt_gray_lvl);
    viewer_part.addPointCloud (cloud_processed, cloud_in_color_h, "cloud_in_v1", v1);

    pcl::visualization::PointCloudColorHandlerCustom<pcl::PointXYZ> cloud_tr_color_h (cloud_objectonly, 20, 180, 20);
    viewer_part.addPointCloud (cloud_objectonly, cloud_tr_color_h, "cloud_tr_v1", v2);

    viewer_part.addText ("The Fucking Original Point Cloud", 10, 15, 16, txt_gray_lvl, txt_gray_lvl, txt_gray_lvl, "icp_info_1", v1);
    viewer_part.addText ("The Fucking Processed Point Cloud", 10, 15, 16, txt_gray_lvl, txt_gray_lvl, txt_gray_lvl, "icp_info_2", v2);

    viewer_part.setBackgroundColor (bckgr_gray_level, bckgr_gray_level, bckgr_gray_level, v1);
    viewer_part.setBackgroundColor (bckgr_gray_level, bckgr_gray_level, bckgr_gray_level, v2);

    viewer_part.setCameraPosition (-3.68332, 2.94092, 5.71266, 0.289847, 0.921947, -0.256907, 0);
    viewer_part.setSize (1280, 1024);

    while (!viewer_part.wasStopped())
    {
        viewer_part.spinOnce();
    }

    
    // 画出点云和法线
    boost::shared_ptr<pcl::visualization::PCLVisualizer> viewer_normal(new pcl::visualization::PCLVisualizer("3D Viewer"));
    viewer_normal->setBackgroundColor(0, 0, 0);
    viewer_normal->addPointCloud<pcl::PointXYZRGB>(colored_cloud, "sample cloud");
    viewer_normal->setPointCloudRenderingProperties(pcl::visualization::PCL_VISUALIZER_POINT_SIZE, 3, "sample cloud");
    viewer_normal->addPointCloudNormals<pcl::PointXYZRGB, pcl::Normal>(colored_cloud, normals, 10, 0.2, "normals");
    viewer_normal->addCoordinateSystem(1.0);
    viewer_normal->initCameraParameters();
    viewer_normal->setCameraPosition (-3.68332, 2.94092, 5.71266, 0.289847, 0.921947, -0.256907, 0);
    while (!viewer_normal->wasStopped())
    {
        viewer_normal->spinOnce(100);
        boost::this_thread::sleep(boost::posix_time::microseconds(100000));
    }
}

pcl::PointCloud<pcl::PointXYZ>::Ptr loadPointCloud(std::string path)
{

    pcl::PointCloud<pcl::PointXYZ>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZ>);
    if (pcl::io::loadPLYFile<pcl::PointXYZ>(path, *cloud) == -1) 
    {       
        PCL_ERROR("Couldnot read file.\n");
        return 0;
    }
    std::cout<<"pointcloud size: "<<cloud->width<<" * "<<cloud->height << std::endl; 
    return cloud;
}

int main(int argc, char** argv)
{
    pcl::PointCloud<pcl::PointXYZ>::Ptr cloud;
    cloud = loadPointCloud("../standard.ply");

    regionGrowing(cloud);

	return 1;
}

附赠的CMakeLists.txt

cmake_minimum_required(VERSION 2.8.7)
project(test)

find_package(PCL 1.5 REQUIRED)

include_directories(${PCL_INCLUDE_DIRS})
link_directories(${PCL_LIBRARY_DIRS})
add_definitions(${PCL_DEFINITIONS})

set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -std=c++11 -fexceptions -frtti -pthread -O3 -march=core2")

set(ROOT 		"${CMAKE_CURRENT_SOURCE_DIR}/")

include_directories(
    ${ROOT}
    ${ROOT}/include
)

file(GLOB SOURCES
    "*.cpp"
    )

link_directories(
    ${ROOT}/lib
)

add_executable(test ${SOURCES})
target_link_libraries(test ${PCL_LIBRARIES})

标签：gray,PCL,viewer,C++,pcl,点云,lvl,txt,cloud
来源： https://blog.csdn.net/weixin_42156097/article/details/111474829