关键点简介
关键点也称为兴趣点,它是 2D 图像或 3D 点云或曲面模型上,可以通过检测标准来获取的具有稳定性、区别性的点集。从技术上来说,关键点的数量比原始点云或图像的数据量少很多,其与局部特征描述子结合组成关键点描述子。常用来构成原始数据的紧凑表示,具有代表性与描述性,从而加快后续识别、追踪等对数据的处理速度 。(也就是能够代表对象特征的子点集)
固而,关键点提取就成为了2D与3D信息处理中不可获缺的关键技术
关键点概念以及算法
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,对NARF关键点提取过程有以下要求:
- 提取的过程必须考虑边缘以及物体表面变化信息
- 即使换了不同的视角,关键点的位置必须稳定的可以被重复探测
- 关键点所在的位置必须有稳定的支持区域,可以计算描述子和估计唯一的法向量。
为了满足上述要求,可以通过以下探测步骤来进行关键点提取:
- 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
- 遍历每个深度图像点,根据近邻区域的表面变化决定一测度表面变化的系数,以及变化的主方向;
- 根据第2步找到的主方向计算兴趣值,表征该方向与其他方向的不同,以及该处表面的变化情况,即该点有多稳定;
- 对兴趣值进行平滑过滤;
- 进行非极大值抑制找到最终的关键点,即为 NARF 关键点。
- 实践项目
narf_keypoint_extraction.cpp
#include <iostream>
#include <boost/thread/thread.hpp>
#include <pcl/range_image/range_image.h>
#include <pcl/io/pcd_io.h>
#include <pcl/visualization/range_image_visualizer.h>
#include <pcl/visualization/pcl_visualizer.h>
#include <pcl/features/range_image_border_extractor.h>
#include <pcl/keypoints/narf_keypoint.h>
#include <pcl/console/parse.h>
#include <pcl/common/file_io.h>
typedef pcl::PointXYZ PointType;
// --------------------
// -----Parameters-----
// --------------------
float angular_resolution=0.5f; //应该时求解 局部描述子的时候,对应的beam的间隔角度 这个觉得了直方差的维度
float support_size=0.2f; //这个有两个作用,一个是 support_size/2 对应的算术特征点提取的局部参考空间大小,另外一个在求解局部描述子时,对应的patch的覆盖范围
pcl::RangeImage::CoordinateFrame coordinate_frame=pcl::RangeImage::CAMERA_FRAME; //深度世界坐标系的类型
bool setUnseenToMaxRange=false; //对于超出传感器范围的量采取的措施
void printUsage(const char *progName) {
std::cout << "\n\nUsage: " << progName << " [options] <scene.pcd>\n\n"
<< "Options:\n"
<< "-------------------------------------------\n"
<< "-r <float> angular resolution in degrees (default " << angular_resolution << ")\n"
<< "-c <int> coordinate frame (default " << (int) coordinate_frame << ")\n"
<< "-m Treat all unseen points as maximum range readings\n"
<< "-s <float> support size for the interest points (diameter of the used sphere - "
<< "default " << support_size << ")\n"
<< "-h this help\n"
<< "\n\n";
}
// --------------
// -----Main-----
// --------------
int main(int argc,char **argv){
// --------------------------------------
// -----Parse Command Line Arguments-----
// --------------------------------------
// 显示help参数
if (pcl::console::find_argument(argc, argv, "-h") >= 0) {
printUsage(argv[0]);
return 0;
}
// 对待超出距离的点的处理方法,true就是最大化处理
if (pcl::console::find_argument(argc, argv, "-m") >= 0) {
setUnseenToMaxRange = true;
cout << "Setting unseen values in range image to maximum range readings.\n";
}
// 读取坐标轴的类型,这个可以查查api的定义
int tmp_coordinate_frame;
if (pcl::console::parse(argc, argv, "-c", tmp_coordinate_frame) >= 0) {
coordinate_frame = pcl::RangeImage::CoordinateFrame(tmp_coordinate_frame);
cout << "Using coordinate frame " << (int) coordinate_frame << ".\n";
}
//从命令参数里面获取angular_resolution
//从命令参数里面获取support_size
if (pcl::console::parse(argc, argv, "-s", support_size) >= 0)
cout << "Setting support size to " << support_size << ".\n";
if (pcl::console::parse(argc, argv, "-r", angular_resolution) >= 0)
cout << "Setting angular resolution to " << angular_resolution << "deg.\n";
//从角度变换到弧度
angular_resolution = pcl::deg2rad(angular_resolution);
// ------------------------------------------------------------------
// -----Read pcd file or create example point cloud if not given-----
// ------------------------------------------------------------------
// 下面需要完成以下内容,从文件中读取pcd点云
// 通过点云,转变成深度图
// 加载远景处理,如果没有的点,就直接最大化距离就可以了
pcl::PointCloud<PointType>::Ptr point_cloud_ptr(new pcl::PointCloud<PointType>);
pcl::PointCloud<PointType> &point_cloud=*point_cloud_ptr;
pcl::PointCloud<pcl::PointWithViewpoint> far_ranges;
Eigen::Affine3f scene_sensor_pose(Eigen::Affine3f::Identity());
//在命令行参数列表里面获取pcd文件名并加载
std::vector<int> pcd_filename_indices = pcl::console::parse_file_extension_argument(argc, argv, "pcd");
if (!pcd_filename_indices.empty()) {
std::string filename = argv[pcd_filename_indices[0]];
if (pcl::io::loadPCDFile(filename, point_cloud) == -1) {
cerr << "Was not able to open file \"" << filename << "\".\n";
printUsage(argv[0]);
return 0;
}
//从传感器位置里面获取场景传感器参数
scene_sensor_pose = Eigen::Affine3f(Eigen::Translation3f(point_cloud.sensor_origin_[0],
point_cloud.sensor_origin_[1],
point_cloud.sensor_origin_[2])) *
Eigen::Affine3f(point_cloud.sensor_orientation_);
//加载远景处理图,如果没有就不用了
std::string far_ranges_filename = pcl::getFilenameWithoutExtension(filename) + "_far_ranges.pcd";
if (pcl::io::loadPCDFile(far_ranges_filename.c_str(), far_ranges) == -1)
std::cout << "Far ranges file \"" << far_ranges_filename << "\" does not exists.\n";
}else{
setUnseenToMaxRange=true;
cout << "\nNo *.pcd file given => Generating example point cloud.\n\n";
for (float x=-0.5f;x<=0.5f;x+=0.01f){
for(float y=-0.5f;y<=0.5f;y+=0.01f){
PointType point;
point.x=x;
point.y=y;
point.z=2.0f-y;
point_cloud.points.push_back(point);
}
}
point_cloud.width=(int)point_cloud.points.size();
point_cloud.height=1;
}
// -----------------------------------------------
// -----Create RangeImage from the PointCloud-----
// -----------------------------------------------
float nosie_level=0.0;
//最小范围
float min_range=0.0f;
int border_size=1;
std::shared_ptr<pcl::RangeImage> range_image_ptr(new pcl::RangeImage);
pcl::RangeImage &range_image=*range_image_ptr;
// 将点云变成深度图
// border_size=1 边框尺度
// 噪声级别,如果想要更好的去除噪声,就去更高的值
range_image.createFromPointCloud(point_cloud,angular_resolution,
pcl::deg2rad(360.0f),pcl::deg2rad(180.0f),
scene_sensor_pose,
coordinate_frame,
nosie_level,
min_range,
border_size);
// 加载元景图
range_image.integrateFarRanges(far_ranges);
// 对于没有探测到的点的处理办法
if(setUnseenToMaxRange)
range_image.setUnseenToMaxRange();
// --------------------------------------------
// -----Open 3D viewer and add point cloud-----
// --------------------------------------------
pcl::visualization::PCLVisualizer viewer("3D Viewer");
viewer.setBackgroundColor(1, 1, 1);
//颜色处理
pcl::visualization::PointCloudColorHandlerCustom<pcl::PointWithRange> range_image_color_handler(range_image_ptr,
255, 0, 0);
//深度图加载颜色处理
viewer.addPointCloud(range_image_ptr, range_image_color_handler, "range image");
//设定点大小的属性
viewer.setPointCloudRenderingProperties(pcl::visualization::PCL_VISUALIZER_POINT_SIZE, 2, "range image");
//增加显示坐标轴
viewer.addCoordinateSystem (1.0f, "global");
//PointCloudColorHandlerCustom<PointType> point_cloud_color_handler (point_cloud_ptr, 150, 150, 150);
//viewer.addPointCloud (point_cloud_ptr, point_cloud_color_handler, "original point cloud");
//初始化viewer参数
viewer.initCameraParameters();
//setViewerPose (viewer, range_image.getTransformationToWorldSystem ());
// --------------------------
// -----Show range image-----
// --------------------------
// 显示深度图
pcl::visualization::RangeImageVisualizer range_image_widget("Range image");
range_image_widget.showRangeImage(range_image);
// --------------------------------
// -----Extract NARF keypoints-----
// --------------------------------
//边框抽取器
pcl::RangeImageBorderExtractor range_image_border_extrator;
pcl::NarfKeypoint narf_keypoint_detector(&range_image_border_extrator);
narf_keypoint_detector.setRangeImage(&range_image);
//计算NARF可信度时的范围
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;
// 关键点的索引
pcl::PointCloud<int> keypoint_indices;
//获取关键点 索引
narf_keypoint_detector.compute(keypoint_indices);
std::cout << "Found " << keypoint_indices.points.size() << " key points.\n";
// ----------------------------------------------
// -----Show keypoints in range image widget-----
// ----------------------------------------------
//for (size_t i=0; i<keypoint_indices.points.size (); ++i)
//range_image_widget.markPoint (keypoint_indices.points[i]%range_image.width,
//keypoint_indices.points[i]/range_image.width);
// -------------------------------------
// -----Show keypoints in 3D viewer-----
// -------------------------------------
//显示关键点
pcl::PointCloud<pcl::PointXYZ>::Ptr keypoints_ptr(new pcl::PointCloud<pcl::PointXYZ>);
pcl::PointCloud<pcl::PointXYZ> &keypoints=*keypoints_ptr;
keypoints.points.resize(keypoint_indices.points.size());
for(size_t i=0;i<keypoint_indices.points.size();i++)
{
keypoints.points[i].getVector3fMap() = range_image.points[keypoint_indices.points[i]].getVector3fMap();
}
// 颜色处理,与点大小
pcl::visualization::PointCloudColorHandlerCustom<pcl::PointXYZ> keypoints_color_handler(keypoints_ptr, 0, 255, 0);
viewer.addPointCloud<pcl::PointXYZ>(keypoints_ptr, keypoints_color_handler, "keypoints");
viewer.setPointCloudRenderingProperties(pcl::visualization::PCL_VISUALIZER_POINT_SIZE, 5, "keypoints");
//--------------------
// -----Main loop-----
//--------------------
while (!viewer.wasStopped()) {
range_image_widget.spinOnce(); // process GUI events
viewer.spinOnce();
pcl_sleep(0.01);
}
}