导 Kinect2库,opencv库,pcl库

Kinect2驱动安装: https://blog.csdn.net/qq_15204179/article/details/107706758

ndfreenect2.cmake [Kinect2]

# 查找依赖库
FIND_LIBRARY(freenect2_LIBRARY freenect2PATHS ~/freenect2/libNO_DEFAULT_PATH)
SET(freenect2_LIBRARIES ${freenect2_LIBRARY} pthread)# 查找路径
FIND_PATH(freenect2_INCLUDE_DIR libfreenect2/libfreenect2.hppPATHS ~/freenect2/includeNO_DEFAULT_PATH)
SET(freenect2_INCLUDE_DIRS ${freenect2_INCLUDE_DIR})

FindOpenCV.cmake [OpenCV]


# once done, this will define
#OpenCV_FOUND        -  whether there is OpenCV in the prebuilt libraries
#OpenCV_INCLUDE_DIRS -  include directories for OpenCV
#OpenCV_LIBRARY_DIRS -  library directories for OpenCV
#OpenCV_LIBRARIES    -  link this to use OpenCVset(OpenCV_DIR ~/3rdparty/OpenCV-3.4.9)unset(OpenCV_FOUND)MESSAGE(STATUS "OpenCV_DIR TEST  ${PREBUILT_DIR}")set(OpenCV_INCLUDE_DIRS "${OpenCV_DIR}/include" "${OpenCV_DIR}/include/opencv")
set(OpenCV_LIB_COMPONENTS opencv_core;opencv_ml;opencv_calib3d;opencv_highgui;opencv_superres;opencv_photo;opencv_imgcodecs;opencv_features2d;opencv_viz;opencv_flann;opencv_shape;opencv_stitching;opencv_dnn;opencv_videostab;opencv_imgproc;opencv_objdetect;opencv_video;opencv_videoio;opencv_reg;opencv_aruco;opencv_dpm;opencv_xobjdetect;opencv_xfeatures2d;opencv_surface_matching;opencv_plot;opencv_hfs;opencv_tracking;opencv_rgbd;opencv_text;opencv_dnn_objdetect;opencv_face;opencv_optflow;opencv_bgsegm;opencv_bioinspired;opencv_ximgproc;opencv_saliency;opencv_freetype;opencv_stereo;opencv_img_hash;opencv_fuzzy;opencv_ccalib;opencv_line_descriptor;opencv_hdf;opencv_datasets;opencv_phase_unwrapping;opencv_xphoto;opencv_structured_light)find_path(OpenCV_INCLUDE_DIRS NAMES opencv.h HINTS ${OpenCV_DIR}/include NO_SYSTEM_ENVIRONMENT_PATH)set(OpenCV_LIBRARY_DIRS ${OpenCV_DIR}/lib)FOREACH(cvcomponent ${OpenCV_LIB_COMPONENTS})find_library(lib_${cvcomponent} NAMES ${cvcomponent} HINTS ${OpenCV_DIR}/lib NO_DEFAULT_PATH)set(OpenCV_LIBRARIES ${OpenCV_LIBRARIES};${lib_${cvcomponent}})
ENDFOREACH()set(OpenCV_LIBS ${OpenCV_LIBRARIES})set(OpenCV_INCLUDE_DIRS ${OpenCV_INCLUDE_DIRS};${OpenCV_INCLUDE_DIRS}/opencv)if (OpenCV_INCLUDE_DIRS AND OpenCV_LIBRARIES)set(OpenCV_FOUND TRUE)
endif (OpenCV_INCLUDE_DIRS AND OpenCV_LIBRARIES)if (OpenCV_FOUND)if (NOT OpenCV_FIND_QUIETLY)message(STATUS "Found OpenCV: ${OpenCV_LIBRARIES}")endif (NOT OpenCV_FIND_QUIETLY)
else(OpenCV_FOUND)if (OpenCV_FIND_REQUIRED)message(FATAL_ERROR "Could not find the OpenCV library")endif ()
endif (OpenCV_FOUND)

CMakeLists.txt

cmake_minimum_required(VERSION 3.16)
project(kinect2demo)set(CMAKE_CXX_STANDARD 14)set(OUTPUT_DIRECTORY_ROOT ${CMAKE_CURRENT_SOURCE_DIR}/build/debug)
set(CMAKE_RUNTIME_OUTPUT_DIRECTORY "${OUTPUT_DIRECTORY_ROOT}/bin" CACHE PATH "Runtime directory" FORCE)
set(CMAKE_LIBRARY_OUTPUT_DIRECTORY "${OUTPUT_DIRECTORY_ROOT}/lib" CACHE PATH "Library directory" FORCE)
set(CMAKE_ARCHIVE_OUTPUT_DIRECTORY "${OUTPUT_DIRECTORY_ROOT}/lib" CACHE PATH "Archive directory" FORCE)set(CMAKE_MODULE_PATH ${CMAKE_MODULE_PATH} ${CMAKE_SOURCE_DIR}/cmake)
#set(CMAKE_MODULE_PATH ${CMAKE_MODULE_PATH} ~/3rdparty/OpenCV-3.4.9/share/OpenCV)# -------------------------------------------------- 2
find_package(freenect2 REQUIRED)
find_package(OpenCV REQUIRED)
find_package(PCL REQUIRED)include_directories(${freenect2_INCLUDE_DIR})
include_directories(${OpenCV_INCLUDE_DIRS})
include_directories(${PCL_INCLUDE_DIRS})link_directories(${OpenCV_LIBRARY_DIRS})
link_directories(${PCL_LIBRARY_DIRS})add_definitions(${PCL_DEFINITIONS})add_subdirectory(src)
add_subdirectory(src/camera)add_executable(kinect2demo main.cpp)

------------------------------------------------------------------------------------------------------------------------------------------------

01-Kinect2Camera.cpp Kinect2获取保存彩色图及深度图

//
// Created by ty on 20-6-3.
//#include <iostream>
#include <libfreenect2/libfreenect2.hpp>
#include <libfreenect2/packet_pipeline.h>
#include <libfreenect2/frame_listener_impl.h>
#include <libfreenect2/registration.h>#include <opencv2/opencv.hpp>using namespace std;
using namespace cv;const int ACTION_ESC = 27;
const int ACTION_SPACE = 32;// 图片的输出目录
const string output_dir = "../output";/*** 获取Kinect2相机的内容** in:*      Kinect2硬件设备* out:*      彩色图*      深度图**  过滤深度值 去掉无限值|NAN*  水平翻转*  保存彩色图和深度图*  缩放为0.5倍*/
int main(int argc, char *argv[]) {// 创建libfreenect2对象libfreenect2::Freenect2 freenect2;if (freenect2.enumerateDevices() == 0) {std::cerr << "当前未发现Kinect2设备" << std::endl;return -1;}const string &serial = freenect2.getDefaultDeviceSerialNumber();if (serial.empty()) {std::cerr << "设备序列号不存在" << std::endl;return -1;}std::cout << "设备序列号：" << serial << std::endl;// 创建数据处理管道
//    libfreenect2::OpenGLPacketPipeline *pipeline = new libfreenect2::CpuPacketPipeline();libfreenect2::OpenGLPacketPipeline *pipeline = new libfreenect2::OpenGLPacketPipeline();// 打开设备libfreenect2::Freenect2Device *device = freenect2.openDevice(serial, pipeline);// 创建数据接收监听器libfreenect2::SyncMultiFrameListener listener(libfreenect2::Frame::Color | libfreenect2::Frame::Depth);// 给设备设置数据监听器//颜色设备device->setColorFrameListener(&listener);//红外设备  红外与深度是一个相机device->setIrAndDepthFrameListener(&listener);// 启动设备device->start();// 创建对齐工具，帮我们把rgb和depth对齐(去畸变)libfreenect2::Registration *registration = new libfreenect2::Registration(device->getIrCameraParams(),//相机内参,相机中再带存了些;最好自己标定下数据会更准确些device->getColorCameraParams());std::cout << "设备固件版本：" << device->getFirmwareVersion() << std::endl;libfreenect2::Frame undistorted(512, 424, 4),   // 创建对象接收：已经去畸变的深度图,第3个参数:浮点类型的字节数registered(512, 424, 4),    // 创建对象接收：已经对齐的彩色图 （可能会有孔洞）把彩色图对齐到深度图上,第3个参数:浮点类型的字节数bigdepth(1920, 1082, 4);    // 创建对象接收：将小深度图对齐到彩色图上得到的大深度图  1920*1082 float; 1082 第一行与最好一行多了两空像素需要过干掉,第3个参数:浮点类型的字节数// 创建一个容器，准备接收数据libfreenect2::FrameMap frames;int index = 0;while (true) {//用户按下空格保存图片 按下esc退出// 这里会阻塞，直到拿到最新的数据。阻塞才会被释放,必须先释放当前的数据:listener.release(frames);listener.waitForNewFrame(frames);//取颜色图片libfreenect2::Frame *rgb = frames[libfreenect2::Frame::Color];//取深度图片libfreenect2::Frame *depth = frames[libfreenect2::Frame::Depth];//        std::cout << "rgb->width: " << rgb->width << " rgb->height: " << rgb->height << std::endl;
//        std::cout << "depth->width: " << depth->width << " depth->height: " << depth->height << std::endl;// 彩色图与深度图合并; 对原始数据进行对齐 彩色图映射到深度图上  rgb彩色图 ,undistorted去畸变的深度图,registered配准后的彩色图 把彩色图贴到深度图上了 ;是否允许过滤 把相机不可见的像素过滤掉;bigdepth 输出把深度映射到彩色图上的图 大的深度图// bigdepth 输出把深度映射到彩色图上的图 大的深度图 1920*1082 float; 1082 第一行与最好一行多了两空像素需要过干掉registration->apply(rgb, depth, &undistorted, &registered, true, &bigdepth);// 512x424 已和深度对齐的彩色图
//        Mat registeredRgbMat = Mat(registered.height, registered.width, CV_8UC4, registered.data);// 512x424 已去畸变的深度图
//        Mat undistortedDepthMat = Mat(undistorted.height, undistorted.width, CV_32FC1, undistorted.data);// 512x424 未去畸变的深度图
//        Mat smallDepthMat = Mat(depth->height, depth->width, CV_32FC1, depth->data);// 将kinect类型的数据转成OpenCV的Mat//CV_8UC4 u代表无符号 c是通道 通道为4// 1920x1080// opencv提供Mat构造函数 帮我们把unsigned char* 转成 mat对象Mat rgbMat = Mat(rgb->height, rgb->width, CV_8UC4, rgb->data);// 1920x1082  CV_32FC1 32位精度浮点数 单通道  bigDepthMat_ 深度值单位毫米,接收的时候要用CV_32FC1数据类型Mat bigDepthMat_ = Mat(bigdepth.height, bigdepth.width, CV_32FC1, bigdepth.data);//过滤深度值 去掉无限值|NAN, 裁剪//       起始列   起始行       总列数       需要的行数//  Rect_(_Tp _x, _Tp _y, _Tp _width, _Tp _height); 1920*1082 float 截取深度图 获得1080的长度// 截取深度图 获得1080的长度 ,1920*1082 floatMat bigDepthMat = bigDepthMat_(Rect(0, 1, bigDepthMat_.cols, bigDepthMat_.rows - 2));//过滤深度值 去掉无限值|NANfor (int i = 0; i < bigDepthMat.rows; ++i) {for (int j = 0; j < bigDepthMat.cols; ++j) {float &d = bigDepthMat.at<float>(i, j);if (fpclassify(d) == FP_INFINITE || fpclassify(d) == NAN) {d = 0;}}}// *  水平翻转flip(rgbMat, rgbMat, 1);flip(bigDepthMat, bigDepthMat, 1);int action = waitKey(30) & 0xFF;if (action == ACTION_ESC || action == 'q' || action == 'Q') {//按下q 或 esc 27退出break;} else if (action == ACTION_SPACE) {//用户按下空格保存32图片std::cout << "保存rgb图和depth图" << std::endl;
//            *  保存彩色图和深度图// 保存彩图stringstream rr;rr << output_dir << "/rgb_image_" << index << ".jpg";vector<int> params;params.push_back(CV_IMWRITE_JPEG_QUALITY);//图片格式params.push_back(100);//压缩的比例bool rst = imwrite(rr.str(), rgbMat, params);if (!rst) {std::cerr << "目标目录不存在，请检查后再保存： " << output_dir << std::endl;return -1;}std::cout << "保存彩色图成功：" << rr.str() << std::endl;// 保存深度图 深度图全是数字 不是可是的图片,要保存成文件// 将数据转成单通道16位UINT,CV_16UC1 整形16位的保存,接收的时候要用CV_32FC1数据类型bigDepthMat.convertTo(bigDepthMat, CV_16UC1);stringstream dd;dd << output_dir << "/depth_image_" << index << ".xml";FileStorage fs(dd.str(), FileStorage::WRITE);fs << "depth" << bigDepthMat;fs.release();std::cout << "保存深度值成功：" << dd.str() << std::endl;}
//       *  缩放为0.5倍resize(rgbMat, rgbMat, Size(), 0.5, 0.5, INTER_LINEAR);resize(bigDepthMat, bigDepthMat, Size(), 0.5, 0.5, INTER_LINEAR);imshow("color", rgbMat);//彩色图显示imshow("depth", bigDepthMat / 4500.0f);//官方推荐显示深度图除4500//必须先释放当前的数据  waitForNewFrame才会拿到数据listener.release(frames);}device->stop();//停掉设备device->close();//后关闭设备delete registration;//回收数据
}

02-PointCloudMaker.cpp Kinect2转点云图

//
// Created by ty on 20-6-3.
//#include <iostream>#include <opencv2/opencv.hpp>#include <pcl/point_types.h>
#include <pcl/io/pcd_io.h>
#include <pcl/io/io.h>
#include <pcl/visualization/cloud_viewer.h>using namespace std;
using namespace cv;typedef pcl::PointXYZRGB PointT;
typedef pcl::PointCloud<PointT> PointCloud;/*** OpenCV* PCL** 输入：*      彩色图*      深度图*      相机的内参** 输出：*      点云图**/
int main(int argc, char *argv[]) {// 读取彩色图和深度图const Mat &rgbMat = imread("./output1/rgb_image_0.jpg", CV_LOAD_IMAGE_UNCHANGED);// 深度图Mat depthMat;FileStorage fs("./output1/depth_image_0.xml", FileStorage::READ);fs["depth"] >> depthMat;fs.release();// 读取相机内参Mat cameraMatrix = cv::Mat_<double>(3, 3);FileStorage camera_fs("./calib/calibration_in_params.yml", FileStorage::READ);camera_fs["cameraMatrix"] >> cameraMatrix;camera_fs.release();std::cout << cameraMatrix << std::endl;double fx = cameraMatrix.at<double>(0, 0);double fy = cameraMatrix.at<double>(1, 1);double cx = cameraMatrix.at<double>(0, 2);//udouble cy = cameraMatrix.at<double>(1, 2);//v// 准备一个点云，接收数据PointCloud::Ptr cloud(new PointCloud);// 遍历每一个深度图的宽高，给点云的每个点指定其颜色值和深度值for (int v = 0; v < depthMat.rows; ++v) { // 所有行for (int u = 0; u < depthMat.cols; ++u) { // 所有// 取出每个像素对应的Z值ushort d = depthMat.at<ushort>(v, u);//单位毫米double depth_value = double(d) / 1000.0f;//变为米为单位,计算都是用米为单位,存储都是毫米为单位if (depth_value == 0 || isnan(depth_value) || abs(depth_value) < 0.0001) {//depth_value==0 或者不是数字 或者绝对值很小比1微米还小的值continue;}// 创建一个点PointPointT p;// 根据内参计算出其X,Y -----------------------------------------p.z = depth_value;p.x = (u - cx) * p.z / fx;p.y = (v - cy) * p.z / fy;// 给其指定颜色Vec3b bgr = rgbMat.at<Vec3b>(v, u);p.b = bgr[0];p.g = bgr[1];p.r = bgr[2];// 添加到cloud中cloud->points.push_back(p);}}// 显示cloud点云cloud->width = cloud->points.size();//无序点云的宽即是cloud长度cloud->height = 1;cloud->is_dense = false;//是否是密集的pcl::visualization::CloudViewer viewer("cloud_viewer");viewer.showCloud(cloud);//保存成点云文件
//    pcl::io::savePCDFileASCII("output/test_pcd.pcd", cloud);
//    std::cerr << "Saved " << cloud.points.size() << " data points to test_pcd.pcd." << std::endl;while (!viewer.wasStopped()) {//}}

03-PhotoCapture.cpp Kinect2 实时获取点云图

#include <iostream>
#include <opencv2/opencv.hpp>
#include <sstream>
#include <cstdlib>#include <pcl/io/pcd_io.h>
#include <pcl/visualization/cloud_viewer.h>
#include <KinectCamera.h>using namespace std;
using namespace cv;float qnan_ = std::numeric_limits<float>::quiet_NaN();
const char *cameraInCailFile = "./calib/calibration_in_params.yml";
// 获取目标目录
//const char *output_dir = "./scene";
// 制作模板目录
const char *output_dir = "./capture_temp";const int ACTION_ESC = 27;
const int ACTION_SPACE = 32;Eigen::Matrix<float, 1920, 1> colmap;
Eigen::Matrix<float, 1080, 1> rowmap;
const short w = 1920, h = 1080;// 准备数据
void prepareMake3D(const double cx, const double cy,const double fx, const double fy) {
//    const int w = 512;
//    const int h = 424;float *pm1 = colmap.data();float *pm2 = rowmap.data();for (int i = 0; i < w; i++) {*pm1++ = (i - cx + 0.5) / fx;}for (int i = 0; i < h; i++) {*pm2++ = (i - cy + 0.5) / fy;}
}/*** 根据内参，合并RGB彩色图和深度图到点云* @param cloud* @param depthMat* @param rgbMat*/
void getCloud(pcl::PointCloud<pcl::PointXYZRGB>::Ptr &cloud, Mat &depthMat, Mat &rgbMat) {const float *itD0 = (float *) depthMat.ptr();const char *itRGB0 = (char *) rgbMat.ptr();if (cloud->size() != w * h)cloud->resize(w * h);// 获取点云的点集合的指针pcl::PointXYZRGB *itP = &cloud->points[0];bool is_dense = true;// 遍历所有行for (size_t y = 0; y < h; ++y) {// 偏差个数, 跳过y行 y * 1920 个const unsigned int offset = y * w;// 深度图指针加上偏差const float *itD = itD0 + offset;// 彩色图指针加上偏差const char *itRGB = itRGB0 + offset * 4;// 获取指定行的系数const float dy = rowmap(y);// 遍历第y行的所有列for (size_t x = 0; x < w; ++x, ++itP, ++itD, itRGB += 4) {const float depth_value = *itD / 1000.0f;if (!isnan(depth_value) && abs(depth_value) >= 0.0001) {const float rx = colmap(x) * depth_value;const float ry = dy * depth_value;itP->z = depth_value;itP->x = rx;itP->y = ry;itP->b = itRGB[0];itP->g = itRGB[1];itP->r = itRGB[2];} else {itP->z = qnan_;itP->x = qnan_;itP->y = qnan_;itP->b = qnan_;itP->g = qnan_;itP->r = qnan_;is_dense = false;}}}cloud->is_dense = is_dense;
}/** 保存Kenect相机的实时图片*  按enter键保存* */
int main(int argc, char *argv[]) {// 显示RGB图// 显示深度图// 显示合成后的点云图auto *camera = new KinectCamera();
//    cv::VideoCapture capture(0);if (!camera->isOpened()) {std::cout << "无法开启摄像头！" << std::endl;return -1;}Mat cameraMatrix = cv::Mat_<double>(3, 3);FileStorage paramFs(cameraInCailFile, FileStorage::READ);paramFs["cameraMatrix"] >> cameraMatrix;// 内参数据double fx = cameraMatrix.at<double>(0, 0);double fy = cameraMatrix.at<double>(1, 1);double cx = cameraMatrix.at<double>(0, 2);double cy = cameraMatrix.at<double>(1, 2);prepareMake3D(cx, cy, fx, fy);// 准备点云查看工具pcl::visualization::PCLVisualizer::Ptr viewer(new pcl::visualization::PCLVisualizer("3D Viewer"));// 创建智能指针点云pcl::PointCloud<pcl::PointXYZRGB>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZRGB>());viewer->setBackgroundColor(0, 0, 0);pcl::visualization::PointCloudColorHandlerRGBField<pcl::PointXYZRGB> rgbHandler(cloud);viewer->addPointCloud<pcl::PointXYZRGB>(cloud, rgbHandler, "sample cloud");viewer->setPointCloudRenderingProperties(pcl::visualization::PCL_VISUALIZER_POINT_SIZE, 1, "sample cloud");int index = 0;while (true) {cv::Mat rgbMat, depthMat;camera->capture(rgbMat, depthMat);if (!viewer->wasStopped()) {// 将RGB和深度信息合成为一个点云图，保存到cloud中getCloud(cloud, depthMat, rgbMat);pcl::visualization::PointCloudColorHandlerRGBField<pcl::PointXYZRGB> rgbHandler(cloud);viewer->updatePointCloud<pcl::PointXYZRGB>(cloud, rgbHandler, "sample cloud");viewer->spinOnce();}int action = cv::waitKey(20) & 0xFF;if (action == ACTION_SPACE || action == 'q') {// 保存点云std::cout << "保存当前的点云" << std::endl;stringstream rr;rr << output_dir << "/camera_rgb_" << index << ".jpg";vector<int> compression_params;compression_params.push_back(CV_IMWRITE_JPEG_QUALITY);  //选择jpegcompression_params.push_back(100); //在这个填入你要的图片质量bool rst = cv::imwrite(rr.str(), rgbMat, compression_params);if (!rst) {std::cerr << "保存失败，请检查该文件夹是否存在：" << output_dir << std::endl;return -1;}cout << rr.str() << endl;depthMat.convertTo(depthMat, CV_16UC1);// 保存深度图stringstream ss;ss << output_dir << "/camera_depth_" << index << ".xml";FileStorage fs(ss.str(), FileStorage::WRITE);fs << "depth" << depthMat;fs.release();stringstream ss_pcd;ss_pcd << output_dir << "/table_scene_" << index << ".pcd";// 保存模板点云图pcl::io::savePCDFile(ss_pcd.str(), *cloud);index++;cout << "成功保存RGB, 深度图, 点云图" << endl;}else if(action == ACTION_ESC) {// 退出break;}cv::imshow("rgb", rgbMat);cv::imshow("depth", depthMat / 4500);}camera->release();return 0;
}

===========================================================================================

Kinect_封装好的驱动

只能获取彩色图

workspas/HandEyeCalibration/cmake/Findfreenect2.cmake

# 查找依赖库
# 头文件include: freenect2_INCLUDE_DIRS
# 依赖库：       freenect2_LIBRARIESFIND_LIBRARY(freenect2_LIBRARY freenect2PATHS ~/freenect2/libNO_DEFAULT_PATH)
SET(freenect2_LIBRARIES ${freenect2_LIBRARY} pthread)# 查找路径
FIND_PATH(freenect2_INCLUDE_DIR libfreenect2/libfreenect2.hppPATHS ~/freenect2/includeNO_DEFAULT_PATH)
SET(freenect2_INCLUDE_DIRS ${freenect2_INCLUDE_DIR})

workspas/HandEyeCalibration/src/camera/CMakeLists.txt

add_library(kinect_camera KinectCamera.cpp KinectCamera.h)
target_link_libraries(kinect_camera ${OpenCV_LIBRARIES} ${freenect2_LIBRARIES})

workspas/HandEyeCalibration/src/camera/KinectCamera.cpp

//
// Created by ty on 20-9-26.
//#include "KinectCamera.h"int KinectCamera::init() {// 判断已连接的设备个数if (freenect2.enumerateDevices() == 0) {std::cerr << "未发现Kinect2设备" << std::endl;return -1;}const string &serial = freenect2.getDefaultDeviceSerialNumber();if (serial.empty()) {std::cerr << "设备序列号不存在" << std::endl;return -1;}std::cout << "设备序列号: " << serial << std::endl;// 创建pipeline接收并处理数据
//    auto *pipeline = new libfreenect2::CpuPacketPipeline();auto pipeline = new libfreenect2::OpenGLPacketPipeline();// 根据序列号开启设备，并设定数据处理通道device = freenect2.openDevice(serial, pipeline);// 创建数据接收器（监听器），给设备执行监听;device->setColorFrameListener(&listener);device->setIrAndDepthFrameListener(&listener);device->start();return 0;
}void KinectCamera::capture(Mat& outputMat) {listener.waitForNewFrame(frames);libfreenect2::Frame *rgb = frames[libfreenect2::Frame::Color];libfreenect2::Frame *depth = frames[libfreenect2::Frame::Depth];Mat colorMat = Mat(rgb->height, rgb->width, CV_8UC4, rgb->data);// 图像进行镜像处理flip(colorMat, colorMat, 1);colorMat.copyTo(outputMat);listener.release(frames);}void KinectCamera::release() {device->stop();device->close();
}KinectCamera::~KinectCamera() {}

workspas/HandEyeCalibration/src/camera/KinectCamera.h

//
// Created by ty on 20-9-26.
//#ifndef HANDEYECALIBRATION_KINECTCAMERA_H
#define HANDEYECALIBRATION_KINECTCAMERA_H#include <iostream>
#include <opencv2/opencv.hpp>
#include <libfreenect2/libfreenect2.hpp>
#include <libfreenect2/packet_pipeline.h>
#include <libfreenect2/frame_listener_impl.h>using namespace std;
using namespace cv;class KinectCamera {private:int status = -2; // -2 初始状态， -1 出现错误， 0 正常可用// 1. 初始化Kinect2相机对象libfreenect2::Freenect2 freenect2;libfreenect2::Freenect2Device *device = nullptr;libfreenect2::SyncMultiFrameListener listener;libfreenect2::FrameMap frames;public:KinectCamera(): listener(libfreenect2::Frame::Color | libfreenect2::Frame::Depth) {status = init();}virtual ~KinectCamera();// 判断设备是否开启bool isOpened() {return status == 0;}virtual int init();virtual void release();void capture(Mat& colorMat);void capture(Mat& colorMat, Mat& depthMat);
};#endif //HANDEYECALIBRATION_KINECTCAMERA_H

使用:

#include <libfreenect2/libfreenect2.hpp>
#include <libfreenect2/packet_pipeline.h>
#include <libfreenect2/frame_listener_impl.h>
#include <opencv2/opencv.hpp>int main(int argc, char *argv[]) {KinectCamera* camera = new KinectCamera();if (!camera->isOpened()) {std::cerr << "相机打开失败" << std::endl;return -1;}while (true) {//  2. 循环接收相机数据rgb，depth (一定要在下次wait之前释放掉frames.   //listener.release(frames);)Mat colorMat, srcMat, gray;camera->capture(colorMat);// 备份一份原图colorMat.copyTo(srcMat);// 3. 在rgb彩色图里查找角点cvtColor(colorMat, gray, COLOR_BGR2GRAY);imageSize = gray.size();}cv::destroyAllWindows();// 设备使用完毕后，及时关闭camera->release();}

------------------------------------------------------------------------------------------------------------------------------------

控制台命令启动相机

cd ~/3rdparty/source/libfreenect2-master/build/bin
./Protonect

导 Kinect2库,opencv库,pcl库相关推荐

3D视觉学习计划之PCL库的基础知识
3D视觉学习计划之PCL库的基础知识一.PCL库的概述 PCL是一个大型跨平台开源C++编程库,它在吸收了前人点云相关研究基础上建立起来,实现了大量点云相关的通用算法和高效数据结构,涉及到点云获取. ...
PCL库官方教程01
PCL 库概览介绍 PCL 库的组件,简短说明各个模块的功能及其之间的交互. 概述 PCL 被拆分为多个模块化库.主要模块如下图所示: Filters 滤波器下图给出了一个去除噪声的例子.由于测量 ...
随笔：送给初次使用PCL库的小伙伴
写在前面: PCL库,之前在使用他的时候,只是各种掉库,觉得自己会调库了,根据案例可以跑出来自己想要的结果(比如计算一个点云的边界,使用RANSAC拟合三维直线等)觉得自己就是掌握了,其实并不然,这样 ...
学习PCL库你应该知道的C++特性
要学会PCL首先要对C++进行学习,所以这里我们首先对PCL库的代码中常见的C++的技巧进行整理和概述,并且对其中的难点进行细化讲解.首先我们搞清楚PCL库的文件形式.是一个以CMake构建的项目,库 ...
Ubuntu下Qt中使用pcl库
pcl依赖及安装 1.一般不用到qt或vtk显示点云 ###pcl通过ppa安装步骤 //这样安装,目前默认安装的是pcl1.7.2 sudo add-apt-repository ppa:v-lau ...
PCL Lesson1 :PCL库PCLVisualizer的简单使用
PCL库PCLVisualizer的简单使用. 包括实例化对象,填充点云,静态显示和动态显示 #include <stdio.h> #include <string> #inc ...
基于opencv第三方视觉库，通过内网IP调用手机摄像头，实现人脸识别与图形监测
1. 安装opencv视觉库 OpenCV 是一个开源的计算机视觉库,OpenCV 库用C语言和 C++ 语言编写,可以在 Windows.Linux.Mac OS X 等系统运行.同时也在积极开发 ...
利用PCL库构建Mesh三维模型
从两张任意拍摄的一对图像(得有大部分重合面积)和相机内参矩阵开始,重建出基于Mesh的三维模型,美观又实用,还不赶快学起来.本文也是记录一下自己学习过程,废话较多,请多包涵,主要代码已注释,请自行下载 ...

导 Kinect2库,opencv库,pcl库

导 Kinect2库,opencv库,pcl库

导 Kinect2库,opencv库,pcl库相关推荐

最新文章

热门文章