fastNlMeansDenoising() does not filter out noise

Question

I am trying to remove noise by opencv fastNlMeansDenoising() function. But My output image is same like original noised image.

Input image:

Code:

#include <iostream>

#include <opencv2/opencv.hpp>

#include <opencv2/highgui/highgui.hpp>

#include <opencv2/imgproc/imgproc.hpp>

using namespace std;

using namespace cv;


int main() {

    Mat img = imread("noisy.jpg");

    if (!img.data) {
        cout << "File not found" << endl;
        return -1;
    }

    // first copy the image
    Mat img_gray = img.clone();
    cvtColor(img, img_gray, CV_RGB2GRAY);

    Mat img1;
    //fastNlMeansDenoising(img_gray, img1, 3.0, 7, 21);
    cv::fastNlMeansDenoising(img_gray, img1, 3.0, 7, 21);

    imshow("img1", img1);

    waitKey();

    return 0;
}

Output Image:

I can not see any effect of smoothing. I do not understand reason of it. Please help me to use this function for removing noise. thanks

Answer 1

In OpenCV, the function is defined as follows

void fastNlMeansDenoising(InputArray src, OutputArray dst, float h=3, int templateWindowSize=7, int searchWindowSize=21 )

where

Parameters: src – Input 8-bit 1-channel, 2-channel or 3-channel

image. dst – Output image with the same size and type as src .

templateWindowSize – Size in pixels of the template patch that is used to compute weights. Should be odd. Recommended value 7 pixels

searchWindowSize – Size in pixels of the window that is used to compute weighted average for given pixel. Should be odd. Affect performance linearly: greater

searchWindowsSize - greater denoising time. Recommended value 21 pixels

h – Parameter regulating filter strength. Big h value perfectly removes noise but also removes image details, smaller h value preserves details but also preserves some noise

Therefore, in order to remove noise, I had to increase the filter strength parameter h, big h value perfectly removes noise, but a smaller h value preserves details and also preserve some noise.

So I perfectly removed the noise by using the function like this:

fastNlMeansDenoising(img_gray, img1, 30.0, 7, 21);

Output:

Note: This function's execution time is too slow in debug mode. For a little bit faster execution time, better to run it in release mode.