PCL学习笔记（二）：PCL官方教程学习
- PCD文件制作
- Features
- 表面法线提取
- Keypoints
- 提取NARF关键点
- KdTree
- Range Image
- How to create a range image from a point cloud
- How to extract borders from range images
- Segmentation
- Plane model segmentation
- Cylinder model segmentation
本节学习PCL的官方文档教程，记录学习过程
PCD文件制作
在gazebo中建立仿真环境，通过Realsense获取点云，保存为PCD文件，一个是走廊path_corner.pcd，一个是几何物体objects.pcd
Features
表面法线提取
参考官方文档：Estimating Surface Normals in a PointCloud
#include <ros/ros.h>
#include <pcl/point_types.h>
#include <pcl/io/pcd_io.h>
#include <pcl/features/normal_3d.h>
#include <pcl_conversions/pcl_conversions.h>
#include <sensor_msgs/PointCloud2.h>// 计算表面法线
pcl::PointCloud<pcl::Normal>::Ptr estimate_surface_normals( pcl::PointCloud<pcl::PointXYZ>::Ptr cloud){// Create the normal estimation class, and pass the input dataset to itpcl::NormalEstimation<pcl::PointXYZ, pcl::Normal> ne;ne.setInputCloud (cloud);// 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 datasetspcl::PointCloud<pcl::Normal>::Ptr cloud_normals (new pcl::PointCloud<pcl::Normal>);// Use all neighbors in a sphere of radius 3cmne.setRadiusSearch (0.03);// Compute the featuresne.compute (*cloud_normals);return cloud_normals;
}int main(int argc, char** argv) {// 读取pcd点云文件pcl::PointCloud<pcl::PointXYZ>::Ptr cloud (new pcl::PointCloud<pcl::PointXYZ>);if (pcl::io::loadPCDFile<pcl::PointXYZ>("./data/objects.pcd", *cloud) == -1) {PCL_ERROR("Couldn't read file objects.pcd\n");return(-1);}// 计算法线pcl::PointCloud<pcl::Normal>::Ptr normals (new pcl::PointCloud<pcl::Normal>);normals = estimate_surface_normals(cloud);std::cout<<"normals points size:"<<normals->points.size()<<std::endl;return 0;
}
其他参考博客：
pcl::Normal的定义以及cout
调用pcl计算法向量，并将法向量可视化
PCL：从法线计算到曲率计算并可视化
Keypoints
提取NARF关键点
参考代码：PCL系列——从深度图像(RangeImage)中提取NARF关键点
参考博客：
PCL关键点检测–NARF关键点
点云库PCL学习——NARF关键点
NARF（Normal Aligned Radial Feature）关键点是为了从深度图像中识别物体而提出的，关键点探测的重要一步是减少特征提取时的搜索空间，把重点放在重要的结构上。
对NARF关键点提取过程有以下要求：提取的过程必须将边缘以及物体表面变化信息考虑在内；关键点的位置必须稳定，可以被重复探测，即使换了不同视角；关键点所在的位置必须具有稳定的支持区域，可以计算描述子并进行唯一的法向量估计。
该方法提取步骤如下：
1、遍历每个深度图像点，通过寻找在邻近区域有深度突变的位置进行边缘检测。
2、遍历每个深度图像点，根据邻近区域的表面变化决定一种测度表面变化的系数，以及变化的主方向。
3、根据第二步找到的主方向计算兴趣值，表征该方向与其它方向的不同，以及该处表面的变化情况，即该点有多稳定
4、对兴趣值进行平滑滤波
5、进行无最大值压缩找到最终的关键点，即为NARF关键点
关键代码
//从点云数据，创建深度图像float noise_level = 0.0;float min_range = 0.0f;int border_size = 1;boost::shared_ptr<pcl::RangeImage> range_image_ptr (new pcl::RangeImage); //创建RangeImage对象（指针）pcl::RangeImage& range_image = *range_image_ptr; //引用range_image.createFromPointCloud (point_cloud, angular_resolution, pcl::deg2rad (360.0f), pcl::deg2rad (180.0f),scene_sensor_pose, coordinate_frame, noise_level, min_range, border_size); //从点云创建深度图像range_image.integrateFarRanges (far_ranges); //整合远距离点云if (setUnseenToMaxRange)range_image.setUnseenToMaxRange ();//提取NARF关键点pcl::RangeImageBorderExtractor range_image_border_extractor; //创建深度图像的边界提取器，用于提取NARF关键点pcl::NarfKeypoint narf_keypoint_detector (&range_image_border_extractor); //创建NARF对象narf_keypoint_detector.setRangeImage (&range_image);narf_keypoint_detector.getParameters ().support_size = support_size;//narf_keypoint_detector.getParameters ().add_points_on_straight_edges = true;//narf_keypoint_detector.getParameters ().distance_for_additional_points = 0.5;
运行结果如下，提取了10个特征点
KdTree
A k-d tree, or k-dimensional tree, is a data structure used in computer science for organizing some number of points in a space with k dimensions. It is a binary search tree with other constraints imposed on it. K-d trees are very useful for range and nearest neighbor searches.
KdTree常用于进行范围搜索或K近邻搜索的计算
参考代码：How to use a KdTree to search
#include <pcl/point_cloud.h>
#include <pcl/point_types.h>
#include <pcl/io/pcd_io.h>
#include <pcl/kdtree/kdtree_flann.h>
#include <vector>
#include <iostream>int main(int argc, char** argv) {// 读取pcd点云文件pcl::PointCloud<pcl::PointXYZ>::Ptr cloud (new pcl::PointCloud<pcl::PointXYZ>);if (pcl::io::loadPCDFile<pcl::PointXYZ>("./data/objects.pcd", *cloud) == -1) {PCL_ERROR("Couldn't read file objects.pcd\n");return(-1);}// 构造kd treepcl::KdTreeFLANN<pcl::PointXYZ> kdtree;kdtree.setInputCloud (cloud);// 选取搜索点pcl::PointXYZ searchPoint;searchPoint.x = 0.1;searchPoint.y = 0.1;searchPoint.z = 0.1;// K nearest neighbor searchint K = 10;std::vector<int> pointIdxKNNSearch(K);std::vector<float> pointKNNSquaredDistance(K);std::cout << "K nearest neighbor search at (" << searchPoint.x << " " << searchPoint.y << " " << searchPoint.z<< ") with K=" << K << std::endl;if ( kdtree.nearestKSearch (searchPoint, K, pointIdxKNNSearch, pointKNNSquaredDistance) > 0 ){for (std::size_t i = 0; i < pointIdxKNNSearch.size (); ++i)std::cout << " " << (*cloud)[ pointIdxKNNSearch[i] ].x << " " << (*cloud)[ pointIdxKNNSearch[i] ].y << " " << (*cloud)[ pointIdxKNNSearch[i] ].z << " (squared distance: " << pointKNNSquaredDistance[i] << ")" << std::endl;}// Neighbors within radius searchstd::vector<int> pointIdxRadiusSearch;std::vector<float> pointRadiusSquaredDistance;float radius = 1;std::cout << "Neighbors within radius search at (" << searchPoint.x << " " << searchPoint.y << " " << searchPoint.z<< ") with radius=" << radius << std::endl;if ( kdtree.radiusSearch (searchPoint, radius, pointIdxRadiusSearch, pointRadiusSquaredDistance) > 0 ){for (std::size_t i = 0; i < pointIdxRadiusSearch.size (); ++i)std::cout << " " << (*cloud)[ pointIdxRadiusSearch[i] ].x << " " << (*cloud)[ pointIdxRadiusSearch[i] ].y << " " << (*cloud)[ pointIdxRadiusSearch[i] ].z << " (squared distance: " << pointRadiusSquaredDistance[i] << ")" << std::endl;}return 0;
}
Range Image
How to create a range image from a point cloud
参考代码：How to create a range image from a point cloud
根据输入的PCD文件生成Range Image，并进行可视化，结果如下图所示
How to extract borders from range images
参考代码:
How to extract borders from range images
【点云处理技术之PCL】range image——提取深度图像的边界并可视化
从深度图像中提取三种边界：
object borders：目标最外层的边界
veil points：障碍物边界和阴影边界之间的插值点
shadow border：与遮挡相邻的背景点
关键代码
pcl::RangeImageBorderExtractor border_extractor(&range_image);pcl::PointCloud<pcl::BorderDescription> border_descriptions;border_extractor.compute(border_descriptions);
运行结果
Segmentation
Plane model segmentation
参考代码：Plane model segmentation
参考博客：PCL系列——平面模型分割
完成平面模型的分割
Cylinder model segmentation
参考代码：Cylinder model segmentation
参考博客：
PCL 平面、圆柱 模型分割
利用 PCL 分割平面上的物体
算法流程：
- 直通滤波在Z轴上滤除桌后的点云，得到滤波后的点云cloud_filtered
- 法线估计，输入为cloud_filtered，得到法线估计结果cloud_normals
- 平面分割，使用RANSAC算法进行平面分割，输入为cloud_filtered和cloud_normals，需要设置最大迭代次数、距离阈值等参数，得到平面分割的４个系数coefficients_plane，然后通过extract对输入点云cloud_filtered进行提取
- 反向提取剩余的点云，得到cloud_filtered2，获取法线cloud_normals2
- 圆柱分割，同样使用RANSAC算法，输入为cloud_filtered2和cloud_normals2，设置相应的参数，得到分割系数coefficients_cylinder，通过extract提取圆柱点云
原PCD文件
分割结果
