OpenCV中使用SVD分解与重构

OpenCV中SVD分解函数compute

C++: static void SVD::compute(InputArray src, OutputArray w, OutputArray u, OutputArray vt, int flags=0 )

src – Decomposed matrix
w – Computed singular values
u – Computed left singular vectors
v – Computed right singular vectors
vt – Transposed matrix of right singular values
flags – Opertion flags - see SVD::SVD().

使用示例

#include <opencv.hpp>
using namespace cv;

//参数分别为输入图像，输出图像，压缩比例
void SVDRESTRUCT(const cv::Mat &inputImg, cv::Mat &outputImg, double theratio)
{
	cv::Mat tempt;
	cv::Mat U, W, V;
	inputImg.convertTo(tempt, CV_32FC1);
	cv::SVD::compute(tempt, W, U, V);
	cv::Mat w = Mat::zeros(Size(W.rows, W.rows), CV_32FC1);
	int len = theratio*W.rows;
	for (int i = 0; i < len; ++i)
		w.ptr<float>(i)[i] = W.ptr<float>(i)[0];

	cv::Mat result = U*w*V;
	result.convertTo(outputImg, CV_8UC1);
}

int _tmain(int argc, _TCHAR* argv[])
{
	cv::Mat scrX = imread("1.png",0);
	cv::Mat result;
	SVDRESTRUCT(scrX, result,0.1);
	cv::imshow("1",result);
	waitKey(0);
}

       SVD本身是个 O(N^3)的算法，大数据处理比较慢。
原图如下：

原图重构如下：

10%压缩如下：

1%压缩如下：

转载于:https://www.cnblogs.com/didea/p/6044321.html