【PCL】PCL点云分割之欧式聚类分割

---

欧式聚类算法原理

欧式聚类算法基于点之间的欧式距离来对点云进行聚类。其算法流程如下：
1、初始化两个空集，一个作为备选点的点集，一个作为聚类的结果集。
2、对于每个点，计算其与其他点之间的欧式距离。
3、如果点与其他点的欧式距离小于设定的阈值，则将其加入备选点的点集。
4、从备选点的点集中选择一个点作为种子点，将其加入聚类的结果集。
5、从备选点的点集中移除种子点，并将其与种子点距离小于设定的阈值的点加入聚类的结果集。
6、重复步骤5，直到备选点的点集为空。
7、重复步骤4至步骤6，直到所有点都被聚类。
通过欧式聚类算法，可以将点云数据根据其几何特征进行聚类，从而实现对点云的分割和分析。

示例代码

#include <iostream>
#include <pcl/ModelCoefficients.h>
#include <pcl/io/pcd_io.h>
#include <pcl/point_types.h>
#include <pcl/sample_consensus/method_types.h>//模型定义头文件
#include <pcl/sample_consensus/model_types.h>//随机参数估计方法头文件
#include <pcl/segmentation/sac_segmentation.h>//基于采样一致性分割的类的头文件
#include <pcl/visualization/cloud_viewer.h>
#include <pcl/features/normal_3d.h>
#include <pcl/filters/extract_indices.h>
#include <pcl/segmentation/extract_clusters.h>
#include <pcl/segmentation/conditional_euclidean_clustering.h>//条件欧式聚类

int main()
{
	
	pcl::PointCloud<pcl::PointXYZ>::Ptr cloud_in(new pcl::PointCloud<pcl::PointXYZ>);
	pcl::io::loadPCDFile("./segmentTest.pcd", *cloud_in);

	if (cloud_in->empty())
	{
		std::cout << "input cloud is empty";
		return -1;
	}
	std::vector<pcl::PointIndices> clusters;
	pcl::PointCloud<pcl::PointXYZRGB>::Ptr cloud_out(new pcl::PointCloud<pcl::PointXYZRGB>);
	//Creating the KdTree object for the search method of the extraction
	pcl::console::print_highlight("Segment by Euclidean Cluster Method!\n");
	pcl::search::KdTree<pcl::PointXYZ>::Ptr tree(new pcl::search::KdTree<pcl::PointXYZ>);
	tree->setInputCloud(cloud_in);

	//set parameters for EuclideanClusterExtraction
	pcl::EuclideanClusterExtraction<pcl::PointXYZ> ec;
	ec.setClusterTolerance(0.2);
	ec.setMinClusterSize(500);
	ec.setMaxClusterSize(5000);
	ec.setSearchMethod(tree);
	ec.setInputCloud(cloud_in);
	ec.extract(clusters);
	std::cout << "Clusters Number :" << clusters.size() << std::endl;
	//color cloud for visualization
	if (!clusters.empty()) {
		srand(static_cast<unsigned int> (time(0)));
		std::vector<unsigned char> colors;
		for (int i_segment = 0; i_segment < clusters.size(); i_segment++) {
			colors.push_back(static_cast<unsigned char> (rand() % 256));
			colors.push_back(static_cast<unsigned char> (rand() % 256));
			colors.push_back(static_cast<unsigned char> (rand() % 256));
		}

		cloud_out->width = cloud_in->width;
		cloud_out->height = cloud_in->height;
		cloud_out->is_dense = cloud_in->is_dense;
		for (int i_point = 0; i_point < cloud_in->points.size(); i_point++) {
			pcl::PointXYZRGB point;
			point.x = *(cloud_in->points[i_point].data);
			point.y = *(cloud_in->points[i_point].data + 1);
			point.z = *(cloud_in->points[i_point].data + 2);
			point.r = 50;
			point.g = 50;
			point.b = 50;
			cloud_out->points.push_back(point);
		}

		std::vector<pcl::PointIndices>::const_iterator i_segment;
		int next_color = 0;
		for (i_segment = clusters.begin(); i_segment != clusters.end(); i_segment++) {
			std::vector<int>::const_iterator i_point;
			for (i_point = i_segment->indices.begin(); i_point != i_segment->indices.end(); i_point++) {
				int index = *i_point;
				cloud_out->points[index].r = colors[3 * next_color];
				cloud_out->points[index].g = colors[3 * next_color + 1];
				cloud_out->points[index].b = colors[3 * next_color + 2];
			}
			next_color++;
		}
	}

	//可视化
	pcl::visualization::PCLVisualizer::Ptr viewer(new pcl::visualization::PCLVisualizer("3d viewer"));
	//左边窗口显示输入的点云
	int v1(0);
	viewer->createViewPort(0, 0, 0.5, 1, v1);
	viewer->setBackgroundColor(0, 0, 0, v1);
	pcl::visualization::PointCloudColorHandlerCustom<pcl::PointXYZ> color_in(cloud_in, 255, 255, 255);
	viewer->addPointCloud<pcl::PointXYZ>(cloud_in, color_in, "cloud_in", v1);
	viewer->setPointCloudRenderingProperties(pcl::visualization::PCL_VISUALIZER_POINT_SIZE, 3, "cloud_in", v1);
	//右边的窗口显示分割后的点云
	int v2(0);
	viewer->createViewPort(0.5, 0, 1, 1, v2);
	viewer->setBackgroundColor(0, 0, 0, v2);
	viewer->addPointCloud<pcl::PointXYZRGB>(cloud_out, "cloud_out", v2);
	viewer->setPointCloudRenderingProperties(pcl::visualization::PCL_VISUALIZER_POINT_SIZE, 3, "cloud_out", v2);

	pcl::io::savePCDFile("./segmentResult.pcd",*cloud_out);
	while (!viewer->wasStopped())
	{
		viewer->spinOnce(100);
	}
	return 0;

}

测试结果

原点云

分割后点云