点云处理技术之PCL点云分割算法1——平面模型分割圆柱模型分割和欧式聚类提取(含欧式聚类原理)

Posted 2022-02-05 非晚非晚

tags:

篇首语：本文由小常识网(cha138.com)小编为大家整理，主要介绍了点云处理技术之PCL点云分割算法1——平面模型分割圆柱模型分割和欧式聚类提取(含欧式聚类原理)相关的知识，希望对你有一定的参考价值。

文章目录

1. 平面分割
2. 圆柱分割
3. 欧式聚类分割

1. 平面分割

下列中，先随机创建了z=1.0的随机点，然后改变其中3个点的z值。最后，使用SACMODEL_PLANE平面模型对它进行拟合。

#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>

int main(int argc, char **argv)

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

    // Fill in the cloud data，15个点
    cloud->width = 15;
    cloud->height = 1;
    cloud->points.resize(cloud->width * cloud->height);

    // Generate the data
    for (auto &point : *cloud)//同一个z
    
        point.x = 1024 * rand() / (RAND_MAX + 1.0f);
        point.y = 1024 * rand() / (RAND_MAX + 1.0f);
        point.z = 1.0;
    

    // Set a few outliers，设置离群点，设置z就可以
    (*cloud)[0].z = 2.0;
    (*cloud)[3].z = -2.0;
    (*cloud)[6].z = 4.0;

    std::cerr << "Point cloud data: " << cloud->size() << " points" << std::endl;
    for (const auto &point : *cloud)
        std::cerr << "    " << point.x << " "
                  << point.y << " "
                  << point.z << std::endl;

    pcl::ModelCoefficients::Ptr coefficients(new pcl::ModelCoefficients);
    pcl::PointIndices::Ptr inliers(new pcl::PointIndices);
    // Create the segmentation object
    pcl::SACSegmentation<pcl::PointXYZ> seg;
    // Optional
    seg.setOptimizeCoefficients(true);
    // Mandatory
    seg.setModelType(pcl::SACMODEL_PLANE);//平面模型
    seg.setMethodType(pcl::SAC_RANSAC);//设置随机采样一致性方法类型，平面类型
    seg.setDistanceThreshold(0.01);表示点到估计模型的距离最大值

    seg.setInputCloud(cloud);
    seg.segment(*inliers, *coefficients);

    if (inliers->indices.size() == 0)
    
        PCL_ERROR("Could not estimate a planar model for the given dataset.");
        return (-1);
    

    //平面模型类型，系数
    std::cerr << "Model coefficients: " << coefficients->values[0] << " "
              << coefficients->values[1] << " "
              << coefficients->values[2] << " "
              << coefficients->values[3] << std::endl;

    std::cerr << "Model inliers: " << inliers->indices.size() << std::endl;
  for (size_t i = 0; i < inliers->indices.size (); ++i)
    std::cerr << inliers->indices[i] << "    " << cloud->points[inliers->indices[i]].x << " "
                                               << cloud->points[inliers->indices[i]].y << " "
                                               << cloud->points[inliers->indices[i]].z << std::endl;

    return (0);

输出：

Point cloud data: 15 points
    0.352222 -0.151883 2
    -0.106395 -0.397406 1
    -0.473106 0.292602 1
    -0.731898 0.667105 -2
    0.441304 -0.734766 1
    0.854581 -0.0361733 1
    -0.4607 -0.277468 4
    -0.916762 0.183749 1
    0.968809 0.512055 1
    -0.998983 -0.463871 1
    0.691785 0.716053 1
    0.525135 -0.523004 1
    0.439387 0.56706 1
    0.905417 -0.579787 1
    0.898706 -0.504929 1
Model coefficients: 0 0 1 -1
Model inliers: 12
1    -0.106395 -0.397406 1
2    -0.473106 0.292602 1
4    0.441304 -0.734766 1
5    0.854581 -0.0361733 1
7    -0.916762 0.183749 1
8    0.968809 0.512055 1
9    -0.998983 -0.463871 1
10    0.691785 0.716053 1
11    0.525135 -0.523004 1
12    0.439387 0.56706 1
13    0.905417 -0.579787 1
14    0.898706 -0.504929 1

2. 圆柱分割

下例先使用平面分割出平面，使用的是SACMODEL_NORMAL_PLANE，模型约束平面的法向方向，针对复杂的平面，可以更准确分割出平面点云，与SACMODEL_PLANE不同。最后针对剩余的点云进行圆柱分割。

下列中保存了两份分割点云：平面点云和圆柱点云。

#include <pcl/ModelCoefficients.h>
#include <pcl/io/pcd_io.h>
#include <pcl/point_types.h>
#include <pcl/filters/extract_indices.h>
#include <pcl/filters/passthrough.h>
#include <pcl/features/normal_3d.h>
#include <pcl/sample_consensus/method_types.h>
#include <pcl/sample_consensus/model_types.h>
#include <pcl/segmentation/sac_segmentation.h>

typedef pcl::PointXYZ PointT;

int main(int argc, char **argv)

    // All the objects needed
    pcl::PCDReader reader;//pcd读取
    pcl::PassThrough<PointT> pass;//直通滤波器
    pcl::NormalEstimation<PointT, pcl::Normal> ne;//法线估计对象
    pcl::SACSegmentationFromNormals<PointT, pcl::Normal> seg;//分割对象
    pcl::PCDWriter writer;//pcd写
    pcl::ExtractIndices<PointT> extract;//点提取
    pcl::ExtractIndices<pcl::Normal> extract_normals;//点提取
    pcl::search::KdTree<PointT>::Ptr tree(new pcl::search::KdTree<PointT>());

    // Datasets
    pcl::PointCloud<PointT>::Ptr cloud(new pcl::PointCloud<PointT>);
    pcl::PointCloud<PointT>::Ptr cloud_filtered(new pcl::PointCloud<PointT>);
    pcl::PointCloud<pcl::Normal>::Ptr cloud_normals(new pcl::PointCloud<pcl::Normal>);
    pcl::PointCloud<PointT>::Ptr cloud_filtered2(new pcl::PointCloud<PointT>);
    pcl::PointCloud<pcl::Normal>::Ptr cloud_normals2(new pcl::PointCloud<pcl::Normal>);
    pcl::ModelCoefficients::Ptr coefficients_plane(new pcl::ModelCoefficients), coefficients_cylinder(new pcl::ModelCoefficients);
    pcl::PointIndices::Ptr inliers_plane(new pcl::PointIndices), inliers_cylinder(new pcl::PointIndices);

    // Read in the cloud data
    reader.read("../../pcd/table_scene_mug_stereo_textured.pcd", *cloud);
    std::cerr << "PointCloud has: " << cloud->size() << " data points." << std::endl;

    // Build a passthrough filter to remove spurious NaNs
    //过滤掉z（0，1.5）的点
    pass.setInputCloud(cloud);
    pass.setFilterFieldName("z");
    pass.setFilterLimits(0, 1.5);
    pass.filter(*cloud_filtered);
    std::cerr << "PointCloud after filtering has: " << cloud_filtered->size() << " data points." << std::endl;

    // Estimate point normals，法线估计，要基于法线分割
    ne.setSearchMethod(tree);
    ne.setInputCloud(cloud_filtered);
    ne.setKSearch(50);
    ne.compute(*cloud_normals);

    // Create the segmentation object for the planar model and set all the parameters
    seg.setOptimizeCoefficients(true);
    //模型约束平面的法向方向，针对复杂的平面，可以更准确分割出平面点云，与SACMODEL_PLANE不同
    seg.setModelType(pcl::SACMODEL_NORMAL_PLANE);
    seg.setNormalDistanceWeight(0.1);
    seg.setMethodType(pcl::SAC_RANSAC);//RANSAC方法
    seg.setMaxIterations(100);//最大迭代次数
    seg.setDistanceThreshold(0.03);//与模型的最大距离
    seg.setInputCloud(cloud_filtered);
    seg.setInputNormals(cloud_normals);
    // Obtain the plane inliers and coefficients
    seg.segment(*inliers_plane, *coefficients_plane);
    std::cerr << "Plane coefficients: " << *coefficients_plane << std::endl;

    // Extract the planar inliers from the input cloud
    //提取平面点云
    extract.setInputCloud(cloud_filtered);
    extract.setIndices(inliers_plane);
    extract.setNegative(false);

    // Write the planar inliers to disk
    //将平面点云写入pcd文件
    pcl::PointCloud<PointT>::Ptr cloud_plane(new pcl::PointCloud<PointT>());
    extract.filter(*cloud_plane);
    std::cerr << "PointCloud representing the planar component: " << cloud_plane->size() << " data points." << std::endl;
    writer.write("../../pcd/table_scene_mug_stereo_textured_plane.pcd", *cloud_plane, false);

    // Remove the planar inliers, extract the rest，提取剩余部分
    extract.setNegative(true);
    extract.filter(*cloud_filtered2);
    extract_normals.setNegative(true);
    extract_normals.setInputCloud(cloud_normals);
    extract_normals.setIndices(inliers_plane);
    extract_normals.filter(*cloud_normals2);

    // Create the segmentation object for cylinder segmentation and set all the parameters
    seg.setOptimizeCoefficients(true);
    seg.setModelType(pcl::SACMODEL_CYLINDER);//设置圆柱模型
    seg.setMethodType(pcl::SAC_RANSAC);//RANSAC方法
    seg.setNormalDistanceWeight(0.1);//设置表面法线权重系数
    seg.setMaxIterations(10000);//最大迭代次数
    seg.setDistanceThreshold(0.05);//最大阈值
    seg.setRadiusLimits(0, 0.1);//设置估计出的圆柱模型的半径的范围
    seg.setInputCloud(cloud_filtered2);
    seg.setInputNormals(cloud_normals2);

    // Obtain the cylinder inliers and coefficients
    seg.segment(*inliers_cylinder, *coefficients_cylinder);
    std::cerr << "Cylinder coefficients: " << *coefficients_cylinder << std::endl;

    // Write the cylinder inliers to disk
    extract.setInputCloud(cloud_filtered2);
    extract.setIndices(inliers_cylinder);
    extract.setNegative(false);
    pcl::PointCloud<PointT>::Ptr cloud_cylinder(new pcl::PointCloud<PointT>());
    extract.filter(*cloud_cylinder);
    if (cloud_cylinder->points.empty())
        std::cerr << "Can't find the cylindrical component." << std::endl;
    else
    
        std::cerr << "PointCloud representing the cylindrical component: " << cloud_cylinder->size() << " data points." << std::endl;
        writer.write("../../pcd/table_scene_mug_stereo_textured_cylinder.pcd", *cloud_cylinder, false);
    
    return (0);

原始点云为：

分割出来的圆柱为：

分割出来的平面为：

3. 欧式聚类分割

欧式聚类原理：

找到空间中某点 $p_0$ ，有kdTree找到离它最近的n个点，判断这n个点到 $p_0$ 的距离。将距离小于阈值r的点 $p_1,p_2,p_3,p_4....$ 放在类Q里。
对Q中的剩余点重复以上步骤，并将满足条件的点放入Q。
当Q中再也不能有新的点加入时，则完成搜索。

下面的欧式聚类分割代码逻辑为：

首先对原始点云做下采样，体素大小为0.01.
然后采样点云做平面滤波，使用的是SAC_RANSAC平面滤波，直到总量小于原始的0.3倍之后才不做平面滤波
对平面滤波剩余的点云进行欧式距离分割。

#include <pcl/ModelCoefficients.h>
#include <pcl/point_types.h>
#include <pcl/io/pcd_io.h>
#include <pcl/filters/extract_indices.h>
#include <pcl/filters/voxel_grid.h>
#include <pcl/features/normal_3d.h>
#include <pcl/kdtree/kdtree.h>
#include <pcl/sample_consensus/method_types.h>
#include <pcl/sample_consensus/model_types.h>
#include <pcl/segmentation/sac_segmentation.h>
#include <pcl/segmentation/extract_clusters.h>

int main(int argc, char **argv)

    // Read in the cloud data
    pcl::PCDReader reader;
    pcl::PointCloud<pcl::PointXYZ>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZ>), cloud_f(new pcl::PointCloud<pcl::PointXYZ>);
    reader.read("../../pcd/table_scene_lms400.pcd", *cloud);
    std::cout << "PointCloud before filtering has: " << cloud->size() << " data points." << std::endl; //*

    // Create the filtering object: downsample the dataset using a leaf size of 1cm
    //下采样
    pcl::VoxelGrid<pcl::PointXYZ> vg;
    pcl::PointCloud<pcl::PointXYZ>::Ptr cloud_filtered(new pcl::PointCloud<pcl::PointXYZ>);
    vg.setInputCloud(cloud);
    vg.setLeafSize点云处理技术之PCL滤波器——提取索引的点云(pcl::ExtractIndices)
 PCL：多平面分割
 点云处理技术之open3d第三篇：点云的高级操作篇——点云边界框凸包DBSCAN聚类平面分割和隐点移除
 点云处理技术之PCL滤波器——参数化模型(投影点云，pcl::ProjectInliers)
 28 利用平面对应关系求解RT空间转换矩阵
 PCL基于法线的差异来分割点云