GPU CUDA编程2 opencv GPU 图像叠加 空间转换 阈值处理 直方图均衡化 变形 仿射变换 滤波器 腐蚀膨胀 边缘 特征 匹配 人脸检测 背景去除

于 2018-12-28 19:39:18 发布
阅读量2.5k 收藏 30
点赞数 6
分类专栏： 高性能计算 文章标签： opencv cuda gpu
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/xiaoxiaowenqiang/article/details/85330571
版权

opencv图像读取、显示、视频播放等 opencvGPU接口图像add、sub、颜色空间转换、阈值操作等

1 opencv 图像读取

#include <opencv2/opencv.hpp>
#include <iostream>
using namespace cv;
using namespace std;

int main(int argc, char** argv)
{
 // Read the image 
 Mat img = imread("images/cameraman.tif",0);

 // Check for failure in reading an Image
 if (img.empty()) 
 {
  cout << "Could not open an image" << endl;
  return -1;
 }
// 可视化窗口名
 String win_name = "My First Opencv Program"; 
 // Create a window
 namedWindow(win_name); 
 // Show our image inside the created window.
imshow(win_name, img); 

// Wait for any keystroke in the window 
waitKey(0); 
//destroy the created window
 destroyWindow(win_name); 

 return 0;
}

2创建单通道 256x256 0值 矩阵，黑色图

#include <opencv2/opencv.hpp>
#include <iostream>

using namespace cv;
using namespace std;

int main(int argc, char** argv)
{
 // 创建单通道 256x256 0值 矩阵，黑色图
 Mat img(256, 256, CV_8UC1, Scalar(0)); 
 String win_name = "Blank Image"; 
 namedWindow(win_name); 
 imshow(win_name, img); 
 waitKey(0); 
 destroyWindow(win_name); 
 return 0;
}

3 创建三通道 256x256 blue color 图

#include <opencv2/opencv.hpp>
#include <iostream>

using namespace cv;
using namespace std;

int main(int argc, char** argv)
{
 
 // 通道顺序 bgr
 Mat img(256, 256, CV_8UC3, Scalar(255,0,0)); 
 
 String win_name = "Blank Blue Color Image"; 
 namedWindow(win_name); 
 imshow(win_name, img); 

 waitKey(0); 
 destroyWindow(win_name); 
 return 0;
}

4 opencv显示图形

#include <opencv2/opencv.hpp>
#include <iostream>

using namespace cv;
using namespace std;

int main(int argc, char** argv)
{
 // 3通道 512*512图像
 Mat img(512, 512, CV_8UC3, Scalar(0,0,0)); 
 // 画线条，起点，终点，颜色，粗细
 line(img,Point(0,0),Point(511,511),Scalar(0,255,0),7);
 // 画长方形
 rectangle(img,Point(384,0),Point(510,128),Scalar(255,255,0),5);
 // 画圆
 circle(img,Point(447,63), 63, Scalar(0,0,255), -1);
 // 画椭圆
 ellipse(img,Point(256,256),Point(100,100),0,0,180,255,-1);
 
 // 显示文字
 putText( img, "OpenCV!", Point(10,500), FONT_HERSHEY_SIMPLEX, 3,
           Scalar(255, 255, 255), 5, 8 );
	   
 String win_name = "Blank Blue Color Image"; //Name of the window
 namedWindow(win_name); // Create a window
 imshow(win_name, img); // Show our image inside the created window.

 waitKey(0); // Wait for any keystroke in the window
 destroyWindow(win_name); //destroy the created window

 return 0;
}

5 播放视频文件

#include <opencv2/opencv.hpp>
#include <iostream>
using namespace cv;
using namespace std;
int main(int argc, char* argv[])
{
 // 打开视频文件
 VideoCapture cap("images/rhinos.avi"); 
 // if not success, exit program
 if (cap.isOpened() == false) 
 {
  cout << "Cannot open the video file" << endl;
  return -1;
 }
cout<<"Press Q to Quit" << endl;
String win_name = "First Video";
namedWindow(win_name); 
 while (true)
 {
  Mat frame;
  // read a frame
  bool flag = cap.read(frame); 

  //Breaking the while loop at the end of the video
  if (flag == false) 
  {
   break;
  }
  //display the frame 
  imshow(win_name, frame);// 显示该帧图像
  //Wait for 100 ms and key 'q' for exit
  if (waitKey(100) == 'q')
  {
    break;
  }
 }
destroyWindow(win_name);
return 0;
}

6 打开摄像头，播放拍摄到的图像

#include <opencv2/opencv.hpp>
#include <iostream>

using namespace cv;
using namespace std;

int main(int argc, char* argv[])
{
 //open the Webcam
 VideoCapture cap(0); // 打开0号摄像头
 // if not success, exit program
 if (cap.isOpened() == false)  
 {
  cout << "Cannot open Webcam" << endl;
  return -1;
 }
 //get the frames rate of the video
 double fps = cap.get(CAP_PROP_FPS); 
 cout << "Frames per seconds : " << fps << endl;
cout<<"Press Q to Quit" <<endl;
 String win_name = "Webcam Video";
 namedWindow(win_name); //create a window
 while (true)
 {
  Mat frame;
  bool flag = cap.read(frame); // read a new frame from video 
  //show the frame in the created window
  imshow(win_name, frame);
  if (waitKey(1) == 'q')
  {
      break;
  }
 }
return 0;
}

7 opencv GPU 接口 图像相加

#include <iostream>
#include "opencv2/opencv.hpp"

int main (int argc, char* argv[])
{
    // 读取图像，存储在cpu上
    cv::Mat h_img1 = cv::imread("images/cameraman.tif");
    cv::Mat h_img2 = cv::imread("images/circles.png");
    cv::Mat h_result1;
    
    // 定义GPU mat数据
    cv::cuda::GpuMat d_result1,d_img1, d_img2;
    // cpu上的图像 上传到 GPU中   
    d_img1.upload(h_img1);
    d_img2.upload(h_img2);
    
    // 调用GPU接执行 mat add
    cv::cuda::add(d_img1,d_img2, d_result1);
    
    // 下载结果，GPU结果 到 CPU中
    d_result1.download(h_result1);
    
    //显示结果
    cv::imshow("Image1 ", h_img1);
    cv::imshow("Image2 ", h_img2);
    cv::imshow("Result addition ", h_result1);
    cv::imwrite("images/result_add.png", h_result1);
    cv::waitKey();
    return 0;
}

8opencv GPU 接口 图像相减

#include <iostream>
#include "opencv2/opencv.hpp"

int main (int argc, char* argv[])
{
    //Read Two Images 
    cv::Mat h_img1 = cv::imread("images/cameraman.tif");
    cv::Mat h_img2 = cv::imread("images/circles.png");
    cv::Mat h_result1;
    
    // 定义GPU数据
    cv::cuda::GpuMat d_result1,d_img1, d_img2;
    
    // CPU 到 GPU   
    d_img1.upload(h_img1);
    d_img2.upload(h_img2);
    
    // 调用GPU接执行 mat substract
    cv::cuda::subtract(d_img1, d_img2,d_result1);
    
    // gpu 结果 到 cpu
    d_result1.download(h_result1);
    
    // 显示，保存
    cv::imshow("Image1 ", h_img1);
    cv::imshow("Image2 ", h_img2);
    cv::imshow("Result Subtraction ", h_result1);
    cv::imwrite("images/result_add.png", h_result1);
    cv::waitKey();
    return 0;
}

9 opencv GPU 接口 图像 加权叠加

#include <iostream>
#include "opencv2/opencv.hpp"

int main (int argc, char* argv[])
{
    //Read Two Images 
    cv::Mat h_img1 = cv::imread("images/cameraman.tif");
    cv::Mat h_img2 = cv::imread("images/circles.png");
    cv::Mat h_result1;
    
    //定义GPU数据
    cv::cuda::GpuMat d_result1,d_img1, d_img2;
    
    // 上传 输入
    d_img1.upload(h_img1);
    d_img2.upload(h_img2);
    
    // 图像叠加  d_result1 = 0.7*d_img1 + d_img2*0.3 + 0.0
    cv::cuda::addWeighted(d_img1,0.7,d_img2,0.3,0,d_result1);
    
    // 下载结果
    d_result1.download(h_result1);
    cv::imshow("Image1 ", h_img1);
    cv::imshow("Image2 ", h_img2);
    cv::imshow("Result blending ", h_result1);
    cv::imwrite("images/result_add.png", h_result1);
    cv::waitKey();
    return 0;
}

10 opencv GPU 接口 图像 二进制操作 与 或 异或 非

// bitwise_and是对二进制数据进行“与”操作，即对图像（灰度图像或彩色图像均可）每个像素值进行二进制“与”操作，1&1=1，1&0=0，0&1=0，0&0=0
//bitwise_or是对二进制数据进行“或”操作，即对图像（灰度图像或彩色图像均可）每个像素值进行二进制“或”操作，1|1=1，1|0=0，0|1=0，0|0=0
//bitwise_xor是对二进制数据进行“异或”操作，即对图像（灰度图像或彩色图像均可）每个像素值进行二进制“异或”操作，1^1=0,1^0=1,0^1=1,0^0=0
//bitwise_not是对二进制数据进行“非”操作，即对图像（灰度图像或彩色图像均可）每个像素值进行二进制“非”操作，~1=0，~0=1


#include <iostream>
#include "opencv2/opencv.hpp"

int main (int argc, char* argv[])
{
    cv::Mat h_img1 = cv::imread("images/circles.png");
    //Create Device variables
    cv::cuda::GpuMat d_result1,d_img1;
    cv::Mat h_result1;     
    //Upload Image to device
    d_img1.upload(h_img1);

    cv::cuda::bitwise_not(d_img1,d_result1);// 图像非操作 d_result1 = ~d_img1
    
    //Download result back  to host
    d_result1.download(h_result1);
    cv::imshow("Result inversion ", h_result1);
    cv::imwrite("images/result_inversion.png", h_result1);
    cv::waitKey();
    return 0;
}

11 opencv GPU 接口 图像 颜色空间转换

#include <iostream>
#include "opencv2/opencv.hpp"

int main (int argc, char* argv[])
{
        cv::Mat h_img1 = cv::imread("images/autumn.tif");
        //Define device variables
        cv::cuda::GpuMat d_result1,d_result2,d_result3,d_result4,d_img1;
        //Upload Image to device
        d_img1.upload(h_img1);

        //  GPU 接口 图像 颜色空间转换
        cv::cuda::cvtColor(d_img1, d_result1,cv::COLOR_BGR2GRAY);// 彩色到 灰度
        cv::cuda::cvtColor(d_img1, d_result2,cv::COLOR_BGR2RGB); // bgr彩色 到 RGB
        cv::cuda::cvtColor(d_img1, d_result3,cv::COLOR_BGR2HSV); // bgr彩色 到 HSV
        cv::cuda::cvtColor(d_img1, d_result4,cv::COLOR_BGR2YCrCb);//  bgr彩色 到 YCrCb
        
        cv::Mat h_result1,h_result2,h_result3,h_result4;
        // 下载结果到 cpu
        d_result1.download(h_result1);
        d_result2.download(h_result2);
        d_result3.download(h_result3);
        d_result4.download(h_result4);
        
        // 下载
        cv::imshow("Result in Gray ", h_result1);
        cv::imshow("Result in RGB", h_result2);
        cv::imshow("Result in HSV ", h_result3);
        cv::imshow("Result in YCrCb ", h_result4);
        
        cv::waitKey();
        return 0;
}

12 opencv GPU 接口 图像 阈值操作

#include <iostream>
#include "opencv2/opencv.hpp"

int main (int argc, char* argv[])
{
        cv::Mat h_img1 = cv::imread("images/cameraman.tif", 0);
        //Define device variables
        cv::cuda::GpuMat d_result1,d_result2,d_result3,d_result4,d_result5, d_img1;
        //Upload image on device
        d_img1.upload(h_img1);

        // GPU图像阈值操作，                     阈值， 最大值，
        cv::cuda::threshold(d_img1, d_result1, 128.0, 255.0, cv::THRESH_BINARY);    // 二值化，大于阈值为1
        cv::cuda::threshold(d_img1, d_result2, 128.0, 255.0, cv::THRESH_BINARY_INV);// 二值化，大于阈值为0 
        cv::cuda::threshold(d_img1, d_result3, 128.0, 255.0, cv::THRESH_TRUNC);     // 上截断，大于阈值，截断为阈值
        cv::cuda::threshold(d_img1, d_result4, 128.0, 255.0, cv::THRESH_TOZERO);    // 下截断到0，小于阈值，截断为0
        cv::cuda::threshold(d_img1, d_result5, 128.0, 255.0, cv::THRESH_TOZERO_INV);// 上截断到0,大于阈值，截断为0

        cv::Mat h_result1,h_result2,h_result3,h_result4,h_result5;
        //Copy results back to host
        d_result1.download(h_result1);
        d_result2.download(h_result2);
        d_result3.download(h_result3);
        d_result4.download(h_result4);
        d_result5.download(h_result5);
        cv::imshow("Result Threshhold binary ", h_result1);
        cv::imshow("Result Threshhold binary inverse ", h_result2);
        cv::imshow("Result Threshhold truncated ", h_result3);
        cv::imshow("Result Threshhold truncated to zero ", h_result4);
        cv::imshow("Result Threshhold truncated to zero inverse ", h_result5);
        cv::waitKey();

    return 0;
}

13 opencv cpu 阈值操作性能

#include <iostream>
#include "opencv2/opencv.hpp"
int main (int argc, char* argv[])
{
cv::Mat src = cv::imread("images/cameraman.tif", 0);
cv::Mat result_host1,result_host2,result_host3,result_host4,result_host5;
//Get initial time in miliseconds
int64 work_begin = cv::getTickCount(); // 计时
cv::threshold(src, result_host1, 128.0, 255.0, cv::THRESH_BINARY);
cv::threshold(src, result_host2, 128.0, 255.0, cv::THRESH_BINARY_INV);
cv::threshold(src, result_host3, 128.0, 255.0, cv::THRESH_TRUNC);
cv::threshold(src, result_host4, 128.0, 255.0, cv::THRESH_TOZERO);
cv::threshold(src, result_host5, 128.0, 255.0, cv::THRESH_TOZERO_INV);
//Get time after work has finished     
int64 delta = cv::getTickCount() - work_begin;
//Frequency of timer
double freq = cv::getTickFrequency();
double work_fps = freq / delta;
std::cout<<"Performance of Thresholding on CPU: " <<std::endl;
std::cout <<"Time: " << (1/work_fps) <<std::endl;
std::cout <<"FPS: " <<work_fps <<std::endl;
return 0;
}

14) opencv Gpu 阈值操作性能

#include <iostream>
#include "opencv2/opencv.hpp"

int main (int argc, char* argv[])
{
 cv::Mat h_img1 = cv::imread("images/cameraman.tif", 0);
cv::cuda::GpuMat d_result1,d_result2,d_result3,d_result4,d_result5, d_img1;
//Measure initial time ticks
int64 work_begin = cv::getTickCount(); // 计时
d_img1.upload(h_img1);
cv::cuda::threshold(d_img1, d_result1, 128.0, 255.0, cv::THRESH_BINARY);
cv::cuda::threshold(d_img1, d_result2, 128.0, 255.0, cv::THRESH_BINARY_INV);
cv::cuda::threshold(d_img1, d_result3, 128.0, 255.0, cv::THRESH_TRUNC);
cv::cuda::threshold(d_img1, d_result4, 128.0, 255.0, cv::THRESH_TOZERO);
cv::cuda::threshold(d_img1, d_result5, 128.0, 255.0, cv::THRESH_TOZERO_INV);

cv::Mat h_result1,h_result2,h_result3,h_result4,h_result5;
d_result1.download(h_result1);
d_result2.download(h_result2);
d_result3.download(h_result3);
d_result4.download(h_result4);
d_result5.download(h_result5);
//Measure difference in time ticks
int64 delta = cv::getTickCount() - work_begin;
double freq = cv::getTickFrequency();
//Measure frames per second
double work_fps = freq / delta;
std::cout <<"Performance of Thresholding on GPU: " <<std::endl;
std::cout <<"Time: " << (1/work_fps) <<std::endl;
std::cout <<"FPS: " <<work_fps <<std::endl;
 return 0;
}

opencv GPU接口 直方图均衡化 变形 仿射变换 均值滤波器 高斯核滤波器 索贝尔 拉普拉斯 腐蚀膨胀

1 opencv GPU接口 直方图均衡化

#include <iostream>
#include "opencv2/opencv.hpp"


int main ()
{
    cv::Mat h_img1 = cv::imread("images/cameraman.tif",0);
    cv::cuda::GpuMat d_img1,d_result1;
    d_img1.upload(h_img1);
    
    //  GPU接口 直方图均衡化 
    cv::cuda::equalizeHist(d_img1, d_result1);
    
    cv::Mat h_result1;
    d_result1.download(h_result1);
    cv::imshow("Original Image ", h_img1);
	cv::imshow("Histogram Equalized Image", h_result1);
	cv::imwrite("images/result_inversion.png", h_img1);
	cv::imwrite("images/result_inversion.png", h_result1);
    cv::waitKey();
    return 0;
}

2 opencv 转成HSV空间 单通道 直方图均衡化 再合并

#include <iostream>
#include "opencv2/opencv.hpp"


int main ()
{
    cv::Mat h_img1 = cv::imread("images/autumn.tif");
    cv::Mat h_img2,h_result1;
    cvtColor(h_img1, h_img2, cv::COLOR_BGR2HSV);// 转成HSV空间 
    
    // split分割成 多个 单通道
    std::vector< cv::Mat > vec_channels;
    cv::split(h_img2, vec_channels); 
    
    // 对单个通道进行直方图均衡化
    cv::equalizeHist(vec_channels[2], vec_channels[2]);
    
    //M那个通道 合并
    cv::merge(vec_channels, h_img2); 
      
    //Convert the histogram equalized image from HSV to BGR color space again
    cv::cvtColor(h_img2,h_result1, cv::COLOR_HSV2BGR);
	cv::imshow("Original Image ", h_img1);
	cv::imshow("Histogram Equalized Image", h_result1);
    cv::waitKey();
    return 0;
}

3 opencv GPU接口 图像 变形

#include <iostream>
#include "opencv2/opencv.hpp"
#include <iostream>
#include "opencv2/opencv.hpp"

int main ()
{
    cv::Mat h_img1 = cv::imread("images/cameraman.tif",0);
    cv::cuda::GpuMat d_img1,d_result1,d_result2;
    d_img1.upload(h_img1);
    // 原图像尺寸
    int width= d_img1.cols;
    int height = d_img1.size().height;
    
    // gpu cuda接口 图像变形 
    cv::cuda::resize(d_img1,d_result1,cv::Size(200, 200), cv::INTER_CUBIC);// 变形到固定尺寸，
    cv::cuda::resize(d_img1,d_result2,cv::Size(0.5*width, 0.5*height), cv::INTER_LINEAR);// 缩小一半 
// CV_INTER_NN - 最近-邻居插补
// CV_INTER_LINEAR - 双线性插值（默认方法）
// CV_INTER_AREA - 像素面积相关重采样。当缩小图像时，该方法可以避免波纹的出现。当放大图像时，类似于方法CV_INTER_NN。
// CV_INTER_CUBIC - 双三次插值。

    cv::Mat h_result1,h_result2;
    d_result1.download(h_result1);
    d_result2.download(h_result2);
    
    cv::imshow("Original Image ", h_img1);
    cv::imshow("Resized Image", h_result1);
    cv::imshow("Resized Image 2", h_result2);
    cv::imwrite("Resized1.png", h_result1);
    cv::imwrite("Resized2.png", h_result2);
    cv::waitKey();
    return 0;
}

4 opencv GPU接口 图像 仿射变换

#include <iostream>
#include "opencv2/opencv.hpp"

int main ()
{
	cv::Mat h_img1 = cv::imread("images/cameraman.tif",0);
	cv::cuda::GpuMat d_img1,d_result1,d_result2;
	d_img1.upload(h_img1);
	int cols= d_img1.cols;
	int rows = d_img1.size().height;
	
	// 平移变换，x，水平方向平移70，y垂直方向平移50
	cv::Mat trans_mat = (cv::Mat_<double>(2,3) << 1, 0, 70, 
	                                              0, 1, 50);		      
	cv::cuda::warpAffine(d_img1,d_result1,trans_mat,d_img1.size());
	
	// 旋转变换
	cv::Point2f pt(d_img1.cols/2., d_img1.rows/2.); // 中心点   
	cv::Mat r = cv::getRotationMatrix2D(pt, 45, 1.0);// 旋转45度
	cv::cuda::warpAffine(d_img1, d_result2, r, cv::Size(d_img1.cols, d_img1.rows));
	
	cv::Mat h_result1,h_result2;
	d_result1.download(h_result1);
	d_result2.download(h_result2);
	cv::imshow("Original Image ", h_img1);
	cv::imshow("Translated Image", h_result1);
	cv::imshow("Rotated Image", h_result2);
	cv::imwrite("Translated.png", h_result1);
	cv::imwrite("Rotated.png", h_result2);
	cv::waitKey();
	return 0;
}

5 opencv GPU接口 图像 均值滤波器

#include <iostream>
#include "opencv2/opencv.hpp"
int main ()
{
	cv::Mat h_img1 = cv::imread("images/cameraman.tif",0);
	cv::cuda::GpuMat d_img1,d_result3x3,d_result5x5,d_result7x7;

	d_img1.upload(h_img1);
	// cuda滤波器指针
	cv::Ptr<cv::cuda::Filter> filter3x3,filter5x5,filter7x7;
	// 创建cuda滤波器 BoxFilter 
	filter3x3 = cv::cuda::createBoxFilter(CV_8UC1,CV_8UC1,cv::Size(3,3));
	// 执行滤波器
	filter3x3->apply(d_img1, d_result3x3);
	filter5x5 = cv::cuda::createBoxFilter(CV_8UC1,CV_8UC1,cv::Size(5,5));
	filter5x5->apply(d_img1, d_result5x5);
	filter7x7 = cv::cuda::createBoxFilter(CV_8UC1,CV_8UC1,cv::Size(7,7));
	filter7x7->apply(d_img1, d_result7x7);

	cv::Mat h_result3x3,h_result5x5,h_result7x7;
	d_result3x3.download(h_result3x3);
	d_result5x5.download(h_result5x5);
	d_result7x7.download(h_result7x7);

	cv::imshow("Original Image ", h_img1);
	cv::imshow("Blurred with kernel size 3x3", h_result3x3);
	cv::imshow("Blurred with kernel size 5x5", h_result5x5);
	cv::imshow("Blurred with kernel size 7x7", h_result7x7);
	cv::imwrite("Blurred3x3.png", h_result3x3);
	cv::imwrite("Blurred5x5.png", h_result5x5);
	cv::imwrite("Blurred7x7.png", h_result7x7);

	cv::waitKey();
	return 0;
	}

6 opencv GPU接口 图像 高斯核 滤波器

#include <iostream>
#include "opencv2/opencv.hpp"
int main ()
{
	cv::Mat h_img1 = cv::imread("images/cameraman.tif",0);
	cv::cuda::GpuMat d_img1,d_result3x3,d_result5x5,d_result7x7;

	d_img1.upload(h_img1);
	// cuda滤波器指针
	cv::Ptr<cv::cuda::Filter> filter3x3,filter5x5,filter7x7;
	// 创建cuda滤波器 GaussianFilter 
	filter3x3 = cv::cuda::createGaussianFilter(CV_8UC1,CV_8UC1,cv::Size(3,3),1);
	// 执行滤波器
	filter3x3->apply(d_img1, d_result3x3);
	filter5x5 = cv::cuda::createGaussianFilter(CV_8UC1,CV_8UC1,cv::Size(5,5),1);
	filter5x5->apply(d_img1, d_result5x5);
	filter7x7 = cv::cuda::createGaussianFilter(CV_8UC1,CV_8UC1,cv::Size(7,7),1);
	filter7x7->apply(d_img1, d_result7x7);
	
	cv::Mat h_result3x3,h_result5x5,h_result7x7;
	d_result3x3.download(h_result3x3);
	d_result5x5.download(h_result5x5);
	d_result7x7.download(h_result7x7);
	
	cv::imshow("Original Image ", h_img1);
	cv::imshow("Blurred with kernel size 3x3", h_result3x3);
	cv::imshow("Blurred with kernel size 5x5", h_result5x5);
	cv::imshow("Blurred with kernel size 7x7", h_result7x7);
	cv::imwrite("gBlurred3x3.png", h_result3x3);
	cv::imwrite("gBlurred5x5.png", h_result5x5);
	cv::imwrite("gBlurred7x7.png", h_result7x7);
	cv::waitKey();
	return 0;
}

7 cpu 均值滤波器

#include <iostream>
#include "opencv2/opencv.hpp"

int main ()
{
    cv::Mat h_img1 = cv::imread("images/saltpepper.png",0);
	cv::Mat h_result;
	cv::medianBlur(h_img1,h_result,3);
    cv::imshow("Original Image ", h_img1);
	cv::imshow("Median Blur Result", h_result);
	cv::waitKey();
    return 0;
}

8 GPU 索贝尔滤波器

#include <iostream>
#include "opencv2/opencv.hpp"
int main ()
{
	cv::Mat h_img1 = cv::imread("images/blobs.png",0);
	cv::cuda::GpuMat d_img1,d_resultx,d_resulty,d_resultxy;
	d_img1.upload(h_img1);
	
	cv::Ptr<cv::cuda::Filter> filterx,filtery,filterxy;
	// 索贝尔滤波器
	filterx = cv::cuda::createSobelFilter(CV_8UC1,CV_8UC1,1,0);// x水平方向
	filterx->apply(d_img1, d_resultx);
	
	filtery = cv::cuda::createSobelFilter(CV_8UC1,CV_8UC1,0,1);// y垂直方向
	filtery->apply(d_img1, d_resulty);
	
	cv::cuda::add(d_resultx,d_resulty,d_resultxy);    // 叠加
	
	cv::Mat h_resultx,h_resulty,h_resultxy;
	d_resultx.download(h_resultx);
	d_resulty.download(h_resulty);
	d_resultxy.download(h_resultxy);
	cv::imshow("Original Image ", h_img1);
	cv::imshow("Sobel-x derivative", h_resultx);
	cv::imshow("Sobel-y derivative", h_resulty);
	cv::imshow("Sobel-xy derivative", h_resultxy);
	cv::imwrite("sobelx.png", h_resultx);
	cv::imwrite("sobely.png", h_resulty);
	cv::imwrite("sobelxy.png", h_resultxy);
	cv::waitKey();
	return 0;
}

9 GPU ScharrFilter 滤波器 类似 索贝尔 Sobel

#include <iostream>
#include "opencv2/opencv.hpp"
int main ()
{
	cv::Mat h_img1 = cv::imread("images/blobs.png",0);
	cv::cuda::GpuMat d_img1,d_resultx,d_resulty,d_resultxy;
	d_img1.upload(h_img1);

	cv::Ptr<cv::cuda::Filter> filterx,filtery;
	// 创建滤波器
	filterx = cv::cuda::createScharrFilter(CV_8UC1,CV_8UC1,1,0);
	filterx->apply(d_img1, d_resultx);
	filtery = cv::cuda::createScharrFilter(CV_8UC1,CV_8UC1,0,1);
	filtery->apply(d_img1, d_resulty);

	cv::cuda::add(d_resultx,d_resulty,d_resultxy);   

	cv::Mat h_resultx,h_resulty,h_resultxy;
	d_resultx.download(h_resultx);
	d_resulty.download(h_resulty);
	d_resultxy.download(h_resultxy);
	
	cv::imshow("Original Image ", h_img1);
	cv::imshow("Scharr-x derivative", h_resultx);
	cv::imshow("Scharr-y derivative", h_resulty);
	cv::imshow("Scharr-xy derivative", h_resultxy);
	
	cv::imwrite("scharrx.png", h_resultx);
	cv::imwrite("scharry.png", h_resulty);
	cv::imwrite("scharrxy.png", h_resultxy);
	cv::waitKey();
	return 0;
}

9 GPU 拉普拉斯滤波器

#include <iostream>
#include "opencv2/opencv.hpp"
int main ()
{
	cv::Mat h_img1 = cv::imread("images/blobs.png",0);
	cv::cuda::GpuMat d_img1,d_result1,d_result3;
	d_img1.upload(h_img1);
	cv::Ptr<cv::cuda::Filter> filter1,filter3;
	
	// 创建滤波器
	filter1 = cv::cuda::createLaplacianFilter(CV_8UC1,CV_8UC1,1);
	filter1->apply(d_img1, d_result1);
	filter3 = cv::cuda::createLaplacianFilter(CV_8UC1,CV_8UC1,3);
	filter3->apply(d_img1, d_result3);
	
	cv::Mat h_result1,h_result3;
	d_result1.download(h_result1);
	d_result3.download(h_result3);
	
	cv::imshow("Original Image ", h_img1);
	cv::imshow("Laplacian Filter 1", h_result1);
	cv::imshow("Laplacian Filter 3", h_result3);
	
	cv::imwrite("laplacian1.png", h_result1);
	cv::imwrite("laplacian3.png", h_result3);
	cv::waitKey();
	return 0;
}

9 GPU 腐蚀膨胀操作

#include <iostream>
#include "opencv2/opencv.hpp"
int main ()
{
	cv::Mat h_img1 = cv::imread("images/blobs.png",0);
	cv::cuda::GpuMat d_img1,d_resulte,d_resultd,d_resulto, d_resultc;
	cv::Mat element = cv::getStructuringElement(cv::MORPH_RECT,cv::Size(5,5));// 长方形核	
	d_img1.upload(h_img1);
	
	cv::Ptr<cv::cuda::Filter> filtere,filterd,filtero,filterc;
	
	filtere = cv::cuda::createMorphologyFilter(cv::MORPH_ERODE,CV_8UC1,element);// 腐蚀操作
	filtere->apply(d_img1, d_resulte);
	filterd = cv::cuda::createMorphologyFilter(cv::MORPH_DILATE,CV_8UC1,element);//膨胀操作
	filterd->apply(d_img1, d_resultd);
	filtero = cv::cuda::createMorphologyFilter(cv::MORPH_OPEN,CV_8UC1,element);// 开运算
	filtero->apply(d_img1, d_resulto);
	filterc = cv::cuda::createMorphologyFilter(cv::MORPH_CLOSE,CV_8UC1,element);// 闭运算
	filterc->apply(d_img1, d_resultc);

	cv::Mat h_resulte,h_resultd,h_resulto,h_resultc;
	d_resulte.download(h_resulte);
	d_resultd.download(h_resultd);
	d_resulto.download(h_resulto);
	d_resultc.download(h_resultc);
	cv::imshow("Original Image ", h_img1);
	cv::imshow("Erosion", h_resulte);
	cv::imshow("Dilation", h_resultd);
	cv::imshow("Opening", h_resulto);
	cv::imshow("closing", h_resultc);
	cv::imwrite("erosion7.png", h_resulte);
	cv::imwrite("dilation7.png", h_resultd);
	cv::imwrite("opening7.png", h_resulto);
	cv::imwrite("closing7.png", h_resultc);
	cv::waitKey();
	return 0;
}

opencv GPU接口 边缘检测 orb sift 特征 candy hough mog 级联回归人脸检测

1 蓝色通道图像 边缘检测

#include <iostream>
#include "opencv2/opencv.hpp"

using namespace cv;
using namespace std;

 int main( int argc, char** argv )
 {
 // 打开摄像头
    VideoCapture cap(0); //capture the video from web cam
    if ( !cap.isOpened() )  
    {
         cout << "Cannot open the web cam" << endl;
         return -1;
    }
    while (true)
    {
        Mat frame;
// 读取一帧图像
        bool flag = cap.read(frame); 
         if (!flag) 
        {
             cout << "Cannot read a frame from webcam" << endl;
             break;
        }
// 定义GPU数据等
cuda::GpuMat d_frame, d_frame_hsv,d_intermediate,d_result;
cuda::GpuMat d_frame_shsv[3];
cuda::GpuMat d_thresc[3];
Mat h_result;
d_frame.upload(frame);

// GPU brg 转HSV空间
cuda::cvtColor(d_frame, d_frame_hsv, COLOR_BGR2HSV);

// HSV空间 通道分割
cuda::split(d_frame_hsv, d_frame_shsv);

// 阈值，最大值
// 三通道，阈值二值化，大于阈值为1
cuda::threshold(d_frame_shsv[0], d_thresc[0], 110, 130, THRESH_BINARY);
cuda::threshold(d_frame_shsv[1], d_thresc[1], 50, 255, THRESH_BINARY);
cuda::threshold(d_frame_shsv[2], d_thresc[2], 50, 255, THRESH_BINARY);

// 二值化三通通道与操作
cv::cuda::bitwise_and(d_thresc[0], d_thresc[1],d_intermediate);
cv::cuda::bitwise_and(d_intermediate, d_thresc[2], d_result);

d_result.download(h_result);
imshow("Thresholded Image", h_result); 
imshow("Original", frame); 

        if (waitKey(1) == 'q') 
       {
            break; 
       }
    }
   return 0;
}

2 GPU canny边缘检测

#include <cmath>
#include <iostream>
#include "opencv2/opencv.hpp"

using namespace std;
using namespace cv;
using namespace cv::cuda;
int main()
{
	Mat h_image = imread("images/drawing.JPG",0);
	if (h_image.empty())
	{
	    cout << "can not open image"<< endl;
	    return -1;
	}
	GpuMat d_edge,d_image;
	Mat h_edge;
	d_image.upload(h_image);
	
	// 边缘检测指针
	cv::Ptr<cv::cuda::CannyEdgeDetector> canny_edge = cv::cuda::createCannyEdgeDetector(2.0, 100.0, 3, false);
	// 创建边缘检测器
	canny_edge->detect(d_image, d_edge);
	
	d_edge.download(h_edge);
	imshow("source", h_image);
	imshow("detected edges", h_edge);
	 waitKey(0);

	return 0;
}

3 霍夫线变换 cpu 与 gpu 函数 运行时间对比

#include <cmath>
#include <iostream>
#include "opencv2/opencv.hpp"
using namespace std;
using namespace cv;
using namespace cv::cuda;
int main()
{
    Mat h_image = imread("images/drawing.JPG",0);
    if (h_image.empty())
    {
        cout << "can not open image"<< endl;
        return -1;
    }

    Mat h_edge;
    // 首先需要进行边缘检测
    cv::Canny(h_image, h_edge, 100, 200, 3);

    Mat h_imagec;
    cv::cvtColor(h_edge, h_imagec, COLOR_GRAY2BGR);// 变成3通道，可以显示彩色线条
    
    Mat h_imageg = h_imagec.clone();
    vector<Vec4i> h_lines;
    
    // CPU 霍夫线变换计时 HoughLinesP()===================================
    {
        const int64 start = getTickCount(); // 计时=====
        HoughLinesP(h_edge, h_lines, 1, CV_PI / 180, 50, 60, 5);
        const double time_elapsed = (getTickCount() - start) / getTickFrequency();
        cout << "CPU Time : " << time_elapsed * 1000 << " ms" << endl;
        cout << "CPU FPS : " << (1/time_elapsed) << endl;
    }
    for (size_t i = 0; i < h_lines.size(); ++i)
    {
        // 在图像上 画上检测出来的 直线========
        Vec4i line_point = h_lines[i];
        line(h_imagec, Point(line_point[0], line_point[1]), Point(line_point[2], line_point[3]), Scalar(0, 0, 255), 2, LINE_AA);
    }

    GpuMat d_edge, d_lines;// GPU 中 检测结果
    
    d_edge.upload(h_edge);// 边缘图像
    {
	const int64 start = getTickCount();
        Ptr<cuda::HoughSegmentDetector> hough = cuda::createHoughSegmentDetector(1.0f, (float) (CV_PI / 180.0f), 50, 5);
        hough->detect(d_edge, d_lines);

        const double time_elapsed = (getTickCount() - start) / getTickFrequency();
        cout << "GPU Time : " << time_elapsed * 1000 << " ms" << endl;
       cout << "GPU FPS : " << (1/time_elapsed) << endl;
    }
    vector<Vec4i> lines_g;
    if (!d_lines.empty())
    {
        lines_g.resize(d_lines.cols);
        Mat h_lines(1, d_lines.cols, CV_32SC4, &lines_g[0]);
        d_lines.download(h_lines);// 拷贝检测结果==========
    }
    for (size_t i = 0; i < lines_g.size(); ++i)
    {
        // 在图像上 画上检测出来的 直线========
        Vec4i line_point = lines_g[i];
        line(h_imageg, Point(line_point[0], line_point[1]), Point(line_point[2], line_point[3]), Scalar(0, 0, 255), 2, LINE_AA);
    }

    imshow("source", h_image);
    imshow("detected lines [CPU]", h_imagec);
    imshow("detected lines [GPU]", h_imageg);
    
    imwrite("hough_source.png", h_image);
    imwrite("hough_cpu_line.png", h_imagec);
    imwrite("hough_gpu_line.png", h_imageg);
    
    waitKey(0);

    return 0;
}

4 OPENCV GPU faster特征点检测

#include <iostream>
#include "opencv2/opencv.hpp"
 
using namespace cv;
using namespace std;
 
int main()
{
	Mat h_image = imread( "images/drawing.JPG", 0 );

	// 创建OPENCV GPU faster特征点检测器
	cv::Ptr<cv::cuda::FastFeatureDetector> detector = cv::cuda::FastFeatureDetector::create(100,true,2);
	
	std::vector<cv::KeyPoint> keypoints;// 特征点不用拷贝????
	
	// 上传图像
	cv::cuda::GpuMat d_image;
	d_image.upload(h_image);
	
	// 执行检测
	detector->detect(d_image, keypoints);
	
	// 绘制特征点
	cv::drawKeypoints(h_image,keypoints,h_image);
	//Show detected keypoints
	imshow("Final Result", h_image );
	waitKey(0);
	return 0;
}

5 OPENCV GPU orb特征点检测

#include <iostream>
#include "opencv2/opencv.hpp"
 
using namespace cv;
using namespace std;
int main()
{
	Mat h_image = imread( "images/drawing.JPG", 0 );

	//  OPENCV GPU orb特征点检测
	cv::Ptr<cv::cuda::ORB> detector = cv::cuda::ORB::create();
	std::vector<cv::KeyPoint> keypoints;

	cv::cuda::GpuMat d_image;
	d_image.upload(h_image);

	detector->detect(d_image, keypoints);

	cv::drawKeypoints(h_image,keypoints,h_image);
	imshow("Final Result", h_image );
	waitKey(0);
	return 0;
}

6 OPENCV GPU SURF特征 检测 描述子匹配 单应变换目标图像位置显示

#include <stdio.h>
#include <iostream>
#include "opencv2/opencv.hpp"
#include "opencv2/core.hpp"
#include "opencv2/imgproc.hpp"
#include "opencv2/highgui.hpp"
#include "opencv2/calib3d.hpp"
#include "opencv2/calib3d/calib3d.hpp"
#include "opencv2/features2d.hpp"
#include "opencv2/xfeatures2d.hpp"
#include "opencv2/xfeatures2d/nonfree.hpp"
#include "opencv2/core/cuda.hpp"
#include "opencv2/cudaarithm.hpp"
#include "opencv2/cudafeatures2d.hpp"
#include "opencv2/xfeatures2d/cuda.hpp"


using namespace cv;
using namespace cv::xfeatures2d;
using namespace std;

int main( int argc, char** argv )
{
        // cpu 数据
	Mat h_object_image = imread( "images/object1.jpg", 0 ); // 带皮牌目标图像
	Mat h_scene_image = imread( "images/scene1.jpg", 0 ); // 场景图像，
	
	// gpu 数据
	cuda::GpuMat d_object_image;
	cuda::GpuMat d_scene_image;
	
	cuda::GpuMat d_keypoints_scene, d_keypoints_object;       // GPU 关键点
	vector< KeyPoint > h_keypoints_scene, h_keypoints_object; // CPU 关键点
	
	cuda::GpuMat d_descriptors_scene, d_descriptors_object;   // GPU 描述子
	// 图像 CPU 上传到 GPU
	d_object_image.upload(h_object_image);
	d_scene_image.upload(h_scene_image);

	// SURF_CUDA 检测器
	cuda::SURF_CUDA surf(100);
	// 检测 特征点并提取 对应的 描述子
	surf( d_object_image, cuda::GpuMat(), d_keypoints_object, d_descriptors_object );
	surf( d_scene_image, cuda::GpuMat(), d_keypoints_scene, d_descriptors_scene );
        
	// Brute Force 暴力匹配器
	Ptr< cuda::DescriptorMatcher > matcher = cuda::DescriptorMatcher::createBFMatcher();
	vector< vector< DMatch> > d_matches;
	matcher->knnMatch(d_descriptors_object, d_descriptors_scene, d_matches, 2);// 最近邻 2个

	// 下载关键点
	surf.downloadKeypoints(d_keypoints_scene, h_keypoints_scene);
	surf.downloadKeypoints(d_keypoints_object, h_keypoints_object);
	
	
	std::vector< DMatch > good_matches;
	for (int k = 0; k < std::min(h_keypoints_object.size()-1, d_matches.size()); k++)
	{
		if ( (d_matches[k][0].distance < 0.6*(d_matches[k][1].distance)) &&
				((int)d_matches[k].size() <= 2 && (int)d_matches[k].size()>0) )
		{
		// 最近匹配 < 0.6*次近匹配
			good_matches.push_back(d_matches[k][0]);
		}
	}
        std::cout << "size:" <<good_matches.size();
	
	// 绘制匹配点对
	Mat h_image_result;
	drawMatches( h_object_image, h_keypoints_object, h_scene_image, h_keypoints_scene,
			good_matches, h_image_result, Scalar::all(-1), Scalar::all(-1),
			vector<char>(), DrawMatchesFlags::DEFAULT );

	
	// 找到对应匹配点对 的 图像像素 2d坐标
	std::vector<Point2f> object;
	std::vector<Point2f> scene;
	for (int i = 0; i < good_matches.size(); i++) 
	{
		object.push_back(h_keypoints_object[good_matches[i].queryIdx].pt);
		scene.push_back(h_keypoints_scene[good_matches[i].trainIdx].pt);
	}
	// 计算匹配平面的 单元变换矩阵
	Mat Homo = findHomography(object, scene, RANSAC);
	std::vector<Point2f> corners(4); // 图像四个角点
	std::vector<Point2f> scene_corners(4);
	
	corners[0] = Point(0, 0);// 最上面
	corners[1] = Point(h_object_image.cols, 0);// 目标图像 四点 单应变换后 计算在 场景图像中的位置
	corners[2] = Point(h_object_image.cols, h_object_image.rows);
	corners[3] = Point(0, h_object_image.rows);
	perspectiveTransform(corners, scene_corners, Homo);// 目标点 ====> 单应变换 ====> 场景中的位置
	
	// 画出四条边界线=====
	line(h_image_result, scene_corners[0] + Point2f(h_object_image.cols, 0),scene_corners[1] + Point2f(h_object_image.cols, 0),	Scalar(255, 0, 0), 4);
	line(h_image_result, scene_corners[1] + Point2f(h_object_image.cols, 0),scene_corners[2] + Point2f(h_object_image.cols, 0),Scalar(255, 0, 0), 4);
	line(h_image_result, scene_corners[2] + Point2f(h_object_image.cols, 0),scene_corners[3] + Point2f(h_object_image.cols, 0),Scalar(255, 0, 0), 4);
	line(h_image_result, scene_corners[3] + Point2f(h_object_image.cols, 0),scene_corners[0] + Point2f(h_object_image.cols, 0),Scalar(255, 0, 0), 4);
	
	imshow("Good Matches & Object detection", h_image_result);
	
	waitKey(0);

	
	return 0;
}

7 OPENCV gpu 级联回归 人脸检测

#include "opencv2/objdetect/objdetect.hpp"
#include "opencv2/highgui/highgui.hpp"
#include "opencv2/imgproc/imgproc.hpp"
#include "opencv2/cudaobjdetect.hpp" 
#include <iostream>
#include <stdio.h>
 
using namespace std;
using namespace cv;
 
int main( )
{
    Mat h_image;
    h_image = imread("images/lena_color_512.tif", 0);  
    
    // OPENCV gpu 级联回归 人脸检测
    Ptr<cuda::CascadeClassifier> cascade = cuda::CascadeClassifier::create("haarcascade_frontalface_alt2.xml");
    
    cuda::GpuMat d_image;
    cuda::GpuMat d_buf;
    d_image.upload(h_image);
    
    //cascadeGPU->setMinNeighbors(0);
    //cascadeGPU->setScaleFactor(1.01);
    cascade->detectMultiScale(d_image, d_buf);// 多尺度检测
    
    // 转换检测结果
    std::vector<Rect> detections;
    cascade->convert(d_buf, detections);
    if (detections.empty())
        std::cout << "No detection." << std::endl;
    
    //转换成彩色图
    cvtColor(h_image,h_image,COLOR_GRAY2BGR);
    for(int i = 0; i < detections.size(); ++i)
    {
      // 画矩形框===== 
      rectangle(h_image, detections[i], Scalar(0,255,255), 5);
    }

    imshow("Result image", h_image);
     
waitKey(0);   
return 0;
}

8 OPENCV gpu 级联回归 人脸检测 处理视频

#include <iostream>
#include <opencv2/opencv.hpp>
using namespace cv;
using namespace std;

int main()
{
    VideoCapture cap(0);// 打开摄像头
    if (!cap.isOpened())
    {
        cerr << "Can not open video source";
        return -1;
    }
    
    std::vector<cv::Rect> h_found;// cpu 目标框
    // OPENCV gpu 级联回归 人脸检测
    cv::Ptr<cv::cuda::CascadeClassifier> cascade = 
                  cv::cuda::CascadeClassifier::create("haarcascade_frontalface_alt2.xml");
		  
    cv::cuda::GpuMat d_frame, d_gray, d_found;
    
    while(1)
    {
        Mat frame;
        if ( !cap.read(frame) ) 
	{
            cerr << "Can not read frame from webcam";
            return -1;
        }
        d_frame.upload(frame);// 上传帧图像到GPU
	// 转换成灰度图
        cv::cuda::cvtColor(d_frame, d_gray, cv::COLOR_BGR2GRAY);
        
	// 多尺度人脸检测
        cascade->detectMultiScale(d_gray, d_found);
	
	// 转换结果
        cascade->convert(d_found, h_found);
        
	for(int i = 0; i < h_found.size(); ++i)
	{
	  // 画矩目标形框===== 
          rectangle(frame, h_found[i], Scalar(0,255,255), 5);
	}

        imshow("Result", frame);
        if (waitKey(1) == 'q') {
            break;
        }
    }

    return 0;
}

9 OPENCV gpu 级联回归 人眼检测

#include "opencv2/objdetect/objdetect.hpp"
#include "opencv2/highgui/highgui.hpp"
#include "opencv2/imgproc/imgproc.hpp"
#include "opencv2/cudaobjdetect.hpp" 
#include <iostream>
#include <stdio.h>
 
using namespace std;
using namespace cv;
 
int main( )
{
    Mat h_image;
    h_image = imread("images/lena_color_512.tif", 0);  // 灰度图格式
    
    // OPENCV gpu 级联回归 人脸检测
    Ptr<cuda::CascadeClassifier> cascade = cuda::CascadeClassifier::create("haarcascade_eye.xml");
    
    cuda::GpuMat d_image;
    cuda::GpuMat d_buf;
    d_image.upload(h_image);
    
    //cascadeGPU->setMinNeighbors(0);
    //cascadeGPU->setScaleFactor(1.01);
    cascade->detectMultiScale(d_image, d_buf);
    
    std::vector<Rect> detections;
    cascade->convert(d_buf, detections);// 转换结果
    
    if (detections.empty())
        std::cout << "No detection." << std::endl;
	
    // 转换成彩色图格式，方便绘制彩色框
    cvtColor(h_image,h_image,COLOR_GRAY2BGR);
    for(int i = 0; i < detections.size(); ++i)
    {
      // 画矩目标形框===== 
      rectangle(h_image, detections[i], Scalar(0,255,255), 5);
    }

    imshow("Result image", h_image);
     
waitKey(0);   
return 0;
}

10 gpu 背景减除进行目标检测 背景/前景 检测

#include <iostream>
#include <string>
#include "opencv2/opencv.hpp"

using namespace std;
using namespace cv;
using namespace cv::cuda;
int main()
{
    VideoCapture cap("abc.avi");
    if (!cap.isOpened())
    {
        cerr << "can not open camera or video file" << endl;
        return -1;
    }
    Mat frame;
    cap.read(frame);
    GpuMat d_frame;
    d_frame.upload(frame);
	
    // MOG算法，即高斯混合模型分离算法，全称Gaussian Mixture-based Background/Foreground Segmentation Algorithm
    Ptr<BackgroundSubtractor> mog = cuda::createBackgroundSubtractorMOG();
    GpuMat d_fgmask,d_fgimage,d_bgimage;
    Mat h_fgmask,h_fgimage,h_bgimage;
    mog->apply(d_frame, d_fgmask, 0.01);
    while(1)
    {
        cap.read(frame);
        if (frame.empty())
            break;
        d_frame.upload(frame);// 上传到GPU
	
        int64 start = cv::getTickCount();// 计时
	
        mog->apply(d_frame, d_fgmask, 0.01);// 运动检测
        mog->getBackgroundImage(d_bgimage);// 获取静态背景
	
        double fps = cv::getTickFrequency() / (cv::getTickCount() - start);
        std::cout << "FPS : " << fps << std::endl;
	
        d_fgimage.create(d_frame.size(), d_frame.type());
        d_fgimage.setTo(Scalar::all(0));
        d_frame.copyTo(d_fgimage, d_fgmask);// 按mask拷贝
	
        d_fgmask.download(h_fgmask);
        d_fgimage.download(h_fgimage);
        d_bgimage.download(h_bgimage);
	
        imshow("image", frame);
	
        imshow("foreground mask", h_fgmask);// 前景
        imshow("foreground image", h_fgimage);
        imshow("mean background image", h_bgimage);// 背景
        if (waitKey(1) == 'q')
            break;
    }

    return 0;
}

11 GMG 统计背景图像估计和每像素贝叶斯分割

#include <iostream>
#include <string>
#include "opencv2/opencv.hpp"
#include "opencv2/core.hpp"
#include "opencv2/core/utility.hpp"
#include "opencv2/cudabgsegm.hpp"
#include "opencv2/cudalegacy.hpp"
#include "opencv2/video.hpp"
#include "opencv2/highgui.hpp"

using namespace std;
using namespace cv;
using namespace cv::cuda;

int main(
)
{
    VideoCapture cap("abc.avi");
    if (!cap.isOpened())
    {
        cerr << "can not open video file" << endl;
        return -1;
    }
    Mat frame;
    cap.read(frame);
    GpuMat d_frame;
    d_frame.upload(frame);
    
    // 背景去除
    Ptr<BackgroundSubtractor> gmg = cuda::createBackgroundSubtractorGMG(40);
    GpuMat d_fgmask,d_fgimage,d_bgimage;
    Mat h_fgmask,h_fgimage,h_bgimage;
    
    gmg->apply(d_frame, d_fgmask);
    
    while(1)
    {
        cap.read(frame);
        if (frame.empty())
            break;
        d_frame.upload(frame);
        int64 start = cv::getTickCount();
	
        gmg->apply(d_frame, d_fgmask, 0.01);
	
        double fps = cv::getTickFrequency() / (cv::getTickCount() - start);
        std::cout << "FPS : " << fps << std::endl;
	
        d_fgimage.create(d_frame.size(), d_frame.type());
        d_fgimage.setTo(Scalar::all(0));
        d_frame.copyTo(d_fgimage, d_fgmask);
        d_fgmask.download(h_fgmask);
        d_fgimage.download(h_fgimage);
	
        imshow("image", frame);
        imshow("foreground mask", h_fgmask);
        imshow("foreground image", h_fgimage);
        if (waitKey(30) == 'q')
            break;
    }
    return 0;
}

f

pycuda

在这里插入图片描述

EwenWanW
关注
  • 6
    点赞
  • 踩
  • 30
    收藏
    觉得还不错? 一键收藏
  • 打赏
    打赏
  • 1
    评论
专栏目录
评论 1
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包

打赏作者

EwenWanW

你的鼓励将是我创作的最大动力

¥1 ¥2 ¥4 ¥6 ¥10 ¥20
扫码支付：¥1
获取中
扫码支付

您的余额不足，请更换扫码支付或充值

打赏作者

实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值