将matlab图像svd方法转换为opencv

Question

我想在visual studio中用c ++编写一个带opencv的程序.我的代码遵循matlab代码:

close all
clear all
clc

%reading and converting the image
inImage=imread('pic.jpg');
inImageD=double(inImage);

[U,S,V]=svd(inImageD);

% Using different number of singular values (diagonal of S) to compress and
% reconstruct the image
dispEr = [];
numSVals = [];
for N=5:25:300
  % store the singular values in a temporary var
  C = S;

  % discard the diagonal values not required for compression
  C(N+1:end,:)=0;
  C(:,N+1:end)=0;

  % Construct an Image using the selected singular values
   D=U*C*V';


  % display and compute error
  figure;
  buffer = sprintf('Image output using %d singular values', N)
  imshow(uint8(D));
  title(buffer);
  error=sum(sum((inImageD-D).^2));

  % store vals for display
  dispEr = [dispEr; error];
  numSVals = [numSVals; N];
end

你有什么看法呢？我想将图像保存在文本文件中,并将其从文件中检索到Mat数组中.我把这部分写成如下:

Mat image;
FileStorage read_file("pic_file.txt", FileStorage::READ);
read_file["pic"] >> image;
read_file.release(); 
Mat P;
image.convertTo(P, CV_32FC3,1.0/255);
SVD svda(P); //or SVD::compute(P,W,U,V);

但我有SVD功能的问题,它不起作用.有没有办法计算图像的SVD压缩？
非常感谢.
Vahids.

Answer 1

这是我的代码:

   int main(int argc, char* argv[])
    {
        // Image matrix
        Mat img;
        Mat result;
        //---------------------------------------------
        //
        //---------------------------------------------
        namedWindow("Source Image");

        namedWindow("Result");
        // Load image in grayscale mode
        img=imread("D:\\ImagesForTest\\cat.bmp",0);
        img.convertTo(img,CV_32FC1,1.0/255.0);
        cout << "Source size:" << img.rows*img.cols <<" elements "<< endl;
        // create SVD 
        cv::SVD s;
        // svd result
        Mat w,u,vt;
        // computations ...
        s.compute(img,w,u,vt);

        // collect Sigma matrix (diagonal - is eigen values, other - zeros)
        // we got it in as vector, transform it to diagonal matrix
        Mat W=Mat::zeros(w.rows,w.rows,CV_32FC1);       
        for(int i=0;i<w.rows;i++)
        {
            W.at<float>(i,i)=w.at<float>(i);
        }

        // reduce rank to k
        int k=25;
        W=W(Range(0,k),Range(0,k));
        u=u(Range::all(),Range(0,k));
        vt=vt(Range(0,k),Range::all());

        // Get compressed image
        result=u*W*vt;
        cout << "Result size:" << u.rows*u.cols+k+vt.rows*vt.cols <<" elements "<< endl;
        //---------------------------------------------
        //
        //--------------------------------------------- 
        imshow("Source Image", img);
        imshow("Result", result);
        cvWaitKey(0);
        return 0;
    }

• 来源和结果图像.