【学习日记】win64配置openni的vs2022编译环境

1

在structure官网下载OpenNI的SDK，注意版本对应无误

2

解压后，运行msi文件，安装提示完成安装

3

安装后，环境变量中自动生成了三项内容

4

vs2022中的配置：

我选择了release，x64平台。
需要设置三处地方，如下图所示。

1、VC++——包含目录填写上文提到的环境变量OPENNI_INCLUDE64的路径

2、VC++——库目录填写上文提到的环境变量OPENNI_LIB64的路径

**3、输出——附加依赖项：**填写OpenNI2.lib

5 其他注意点

1、网上有人说要把Redist文件和编译生成的exe放在一个文件夹下，不知道为什么。似乎不进行这步操作也没有影响。保险起见，我也在复制了一份放在里面。

2、程序运行出现如下报错：

Found no files matching 'C:\WINDOWS\SYSTEM32\OpenNI2\Drivers\*.dll'

我直接在System32下自己新建了文件夹，把dll全部复制过去，然后就不会出现这条错误了。

6 测试是否配置成功

test.cpp

#include <opencv2/opencv.hpp>
#include <OpenNI.h>
#include "opencv2/imgproc/types_c.h"
using namespace openni;
using namespace cv;class Grabber
{public:void InitOpenNI();void InitDevice();void InitDepthStream();void InitColorStream();void Run();private:void CapturePsenseDepthFrame();void CapturePsenseColorFrame();cv::Mat ChangeDepthForDisplay(const cv::Mat& mat);openni::Device* device_;openni::VideoStream* depth_stream_;openni::VideoStream* color_stream_;openni::VideoFrameRef* depth_frame_;openni::VideoFrameRef* color_frame_;
};void Grabber::InitOpenNI()
{auto rc = openni::OpenNI::initialize();if (rc != openni::STATUS_OK){printf("Initialize failed\n%s\n", openni::OpenNI::getExtendedError());exit(0);}
}void Grabber::InitDevice()
{device_ = new openni::Device();auto rc = device_->open(openni::ANY_DEVICE);if (rc != openni::STATUS_OK){printf("Couldn't open device\n%s\n", openni::OpenNI::getExtendedError());exit(0);}
}void Grabber::InitDepthStream()
{depth_stream_ = new openni::VideoStream();// Create depth stream from deviceif (device_->getSensorInfo(openni::SENSOR_DEPTH) != nullptr){auto rc = depth_stream_->create(*device_, openni::SENSOR_DEPTH);if (rc != openni::STATUS_OK){printf("Couldn't create depth stream\n%s\n", openni::OpenNI::getExtendedError());exit(0);}}// Get info about depth sensorconst openni::SensorInfo& sensor_info = *device_->getSensorInfo(openni::SENSOR_DEPTH);const openni::Array<openni::VideoMode>& arr = sensor_info.getSupportedVideoModes();// Look for VGA mode in depth sensor and set it for depth streamfor (int i = 0; i < arr.getSize(); ++i){const openni::VideoMode& vmode = arr[i];if (vmode.getPixelFormat() == openni::PIXEL_FORMAT_DEPTH_1_MM &&vmode.getResolutionX() == 640 &&vmode.getResolutionY() == 480){depth_stream_->setVideoMode(vmode);break;}}// Start the depth streamauto rc = depth_stream_->start();if (rc != openni::STATUS_OK){printf("Couldn't start the depth stream\n%s\n", openni::OpenNI::getExtendedError());exit(0);}depth_frame_ = new openni::VideoFrameRef();
}void Grabber::InitColorStream()
{color_stream_ = new openni::VideoStream();if (device_->getSensorInfo(openni::SENSOR_COLOR) != nullptr){auto rc = color_stream_->create(*device_, openni::SENSOR_COLOR);if (rc != openni::STATUS_OK){printf("Couldn't create color stream\n%s\n", openni::OpenNI::getExtendedError());exit(0);}}// Get info about color sensorconst openni::SensorInfo& sensor_info = *device_->getSensorInfo(openni::SENSOR_COLOR);const openni::Array<openni::VideoMode>& arr = sensor_info.getSupportedVideoModes();// Look for VGA mode and set it for color streamfor (int i = 0; i < arr.getSize(); ++i){const openni::VideoMode& vmode = arr[i];if (vmode.getResolutionX() == 640 &&vmode.getResolutionY() == 480){color_stream_->setVideoMode(vmode);break;}}// Note: Doing image registration earlier than this seems to failif (device_->isImageRegistrationModeSupported(openni::IMAGE_REGISTRATION_DEPTH_TO_COLOR)){auto rc = device_->setImageRegistrationMode(openni::IMAGE_REGISTRATION_DEPTH_TO_COLOR);if (rc == openni::STATUS_OK)std::cout << "Depth to color image registration set success\n";elsestd::cout << "Depth to color image registration set failed\n";}else{std::cout << "Depth to color image registration is not supported!!!\n";}// Start color streamauto rc = color_stream_->start();if (rc != openni::STATUS_OK){printf("Couldn't start the depth stream\n%s\n", openni::OpenNI::getExtendedError());exit(0);}color_frame_ = new openni::VideoFrameRef();
}void Grabber::CapturePsenseDepthFrame()
{auto rc = depth_stream_->readFrame(depth_frame_);if (rc != openni::STATUS_OK){printf("Read failed!\n%s\n", openni::OpenNI::getExtendedError());}if (depth_frame_->getVideoMode().getPixelFormat() != openni::PIXEL_FORMAT_DEPTH_1_MM && depth_frame_->getVideoMode().getPixelFormat() != openni::PIXEL_FORMAT_DEPTH_100_UM){printf("Unexpected frame format\n");}// Get pointer to Primesense depth frameopenni::DepthPixel* dev_buf_ptr = (openni::DepthPixel*)depth_frame_->getData();// Copy frame data to OpenCV matcv::Mat depth_mat(depth_frame_->getHeight(), depth_frame_->getWidth(), CV_16U, dev_buf_ptr);cv::Mat disp_mat = ChangeDepthForDisplay(depth_mat);cv::imshow("Depth", disp_mat);
}void Grabber::CapturePsenseColorFrame()
{// Read from stream to frameauto rc = color_stream_->readFrame(color_frame_);if (rc != openni::STATUS_OK){printf("Read failed!\n%s\n", openni::OpenNI::getExtendedError());}// Pointer to Primesense color frameopenni::RGB888Pixel* dev_buf_ptr = (openni::RGB888Pixel*)color_frame_->getData();// Make mat from camera datacv::Mat color_mat(color_frame_->getHeight(), color_frame_->getWidth(), CV_8UC3, dev_buf_ptr);// Convert to BGR format for OpenCVcv::cvtColor(color_mat, color_mat, CV_RGB2BGR);cv::imshow("Color", color_mat);
}void Grabber::Run()
{openni::VideoStream* streams[] = { depth_stream_, color_stream_ };while (true){int readyStream = -1;auto rc = openni::OpenNI::waitForAnyStream(streams, 2, &readyStream, 2000);if (rc != openni::STATUS_OK){printf("Wait failed! (timeout is %d ms)\n%s\n", 2000, openni::OpenNI::getExtendedError());break;}switch (readyStream){case 0:CapturePsenseDepthFrame();break;case 1:CapturePsenseColorFrame();break;default:printf("Unxpected stream\n");}char c = cv::waitKey(10);if ('q' == c)break;}
}cv::Mat Grabber::ChangeDepthForDisplay(const cv::Mat& mat)
{assert(CV_16U == mat.type());const float depth_near = 500;const float depth_far = 5000;const float alpha = 255.0 / (depth_far - depth_near);const float beta = -depth_near * alpha;cv::Mat fmat;mat.convertTo(fmat, CV_32F);for (int r = 0; r < mat.rows; ++r){for (int c = 0; c < mat.cols; ++c){float v = fmat.at<float>(r, c) * alpha + beta;if (v > 255) v = 255;if (v < 0)   v = 0;fmat.at<float>(r, c) = v;}}cv::Mat bmat;fmat.convertTo(bmat, CV_8U);cv::Mat cmat;cv::cvtColor(bmat, cmat, CV_GRAY2BGR);cv::applyColorMap(cmat, cmat, cv::COLORMAP_OCEAN);return cmat;
}int main()
{Grabber grabber;grabber.InitOpenNI();grabber.InitDevice();grabber.InitDepthStream();grabber.InitColorStream();grabber.Run();return 0;
}

运行，没有问题，可以顺利看到摄像头画面。显示depth和color图。

参考资料

https://structure.io/openni
https://documentation.help/OpenNI-2.0/getting_started.html

【学习日记】win64配置openni的vs2022编译环境相关推荐

配置Abaqus2021 + VS2019 + IVF2020编译环境
配置Abaqus2021 + VS2019 + IVF2020编译环境背景介绍研三需要对Abaqus进行UMAT二次开发,虽然不喜欢Fortran,但是还是需要进行安装. 本文介绍的是我第二次安装 ...
c程序语言设定定义域,4.22C语言学习日记：变量的定义域及编译预处理
学习笔记: 变量的存储类型: 静态存储方式:在程序运行期间由系统分配固定存储空间:从程序开始执行到程序结束: 动态存储方式:根据需要进行动态的分配存储空间:从包含该变量定义的函数开始执行至函数执行结束 ...
vscode配置C/C++ windows编译环境。
1. vscode安装c/c++插件在vscode右边侧边栏的插件扩展中安装c/c++调试插件 2. mingw-w64的安装,推荐直接下载离线版安装第一种方式:通过自动安装程序指引安装. 安装程 ...
Windows MinGW配置C、C++编译环境
写在前面的前面:这篇文章vscode和cpp插件版本有点老了,仅供大家参考,最新的和最详细的更新见我的另一篇文章:https://blog.csdn.net/bat67/article/details ...
配置Mingw64和MSYS2编译环境
快速配置MSYS2 下载MSYS2 安装更换国内镜像源安装新库下载MSYS2 MSYS2是一组工具和库,提供一个易于使用的环境,用于构建.安装和运行本地 Windows 软件.下载地址:MSYS ...
linux 编译opencl,OpenCL编译环境配置（VS+Nvidia）
英伟达的显卡首先要下载安装CUDA开发包,可以参考这里的步骤: VS2015编译环境下CUDA安装配置安装好CUDA之后,OpenCL的配置就已经完成了80%了,剩下的工作就是把OpenCL的路 ...
python安装c编译的软件_Notepad++配置C/C++、C#、Java、Python编译环境详细教程
如果只是测试小程序可以用这种方法比较方便,如果对于大程序建议使用专业的IDE. 经常需要写一些小程序来运行,又不想运行Visual Studio.Eclipse这样的环境,而Notepad++是一个 ...
为Notepad++配置C/C++、C#、Java、Python编译环境
本文转自:http://blog.csdn.net/freewaywalker/article/details/8005468 如果只是测试小程序可以用这种方法比较方便,如果对于大程序建议使用专业的 ...
【PHP学习】—apache配置虚拟主机（基于域名）
什么是虚拟主机虚拟主机是指在网络服务器上分出一定的磁盘空间,用户可以租用此部分空间,以供用户放置站点及应用组件,提供必要的数据存放和传输功能. (简而言之就是服务器) 配置基于域名的虚拟主机假设我 ...

【学习日记】win64配置openni的vs2022编译环境

1

2

3

4

5 其他注意点

6 测试是否配置成功

参考资料

【学习日记】win64配置openni的vs2022编译环境相关推荐

最新文章

热门文章