Hossain Md Shakhawat's profile - activity

Another fact is I am adding same amount of noise to different color images. But when calculating MSE, their values are different for different images. Why is that? According to my understanding it should same for all images, if the amount of noise being added is same.

I am getting the same value as yours while considering the three channels.

//For gray level image

for(int y = 0; y < width; y++){ for(int x = 0; x < height;x++){

        difference = (original.at<double>(x,y) - ref.at<double>(x,y));
        sum = sum + difference*difference;
    }
}
mse = (float)sum /(width*height);
printf("MSE =%.3f\n",mse);

I am calculating MSE for M-pixel gray level images MSE=1/M ∑_(j=1)^M [ I_target (j)- I_reference (j) ] ^2

where I_target is the degraded image and I_reference is the original image

I have added noise to color image. But when I am calculating the MSE, it is very low compared to the added noise. For std=20, the MSE is below 200. But I think MSE should be close to square of standard deviation of the added random noise(if std=20 then MSE around 400).

I have added noise to color image in this way. But when I am calculating the MSE, it is very low compared to the added noise. For std=20, the MSE is below 200. But I think MSE should be close to square of std of the random noise(if std=20 then MSE around 400). How to add noise so that the MSE becomes square of the standard deviation of the noise?

Thanks a lot for your prompt update

I got this error while running a code

Thanks, done

I have an image in RGB and I need the Y component of CIE XYZ color space. What is difference between grayscale image and the Y component of CIE XYZ color space?

I need to apply Sobel filter on image and detect the local maximum gradient for each edges direction. Pixels whose gradient are larger than threshold considered as edge pixel. Now the task is to find the width (pixel width) between the local maximum and local minimum for edge (edge pixels).

This solved my problem

Thanks it fixed the error but the resulting noisy image is some how different. It is a yellowish image. And this is not my desired target. Is my way is wrong ? I am very new at opencv and image processing

Please suggest is there any better and simple way to add noise to colour with varying std of gaussian noise.

This is the error OpenCV Error: Assertion failed (mv[i].size == mv[0].size && mv[i].depth() == depth) in merge, file /Users/hossainmdshakhawat/opencv-3.0.0/modules/core/src/convert.cpp, line 950 libc++abi.dylib: terminating with uncaught exception of type cv::Exception: /Users/hossainmdshakhawat/opencv-3.0.0/modules/core/src/convert.cpp:950: error: (-215) mv[i].size == mv[0].size && mv[i].depth() == depth in function merge

my code:

Mat ycrcb = input.clone(); cvtColor(input, ycrcb, CV_BGR2YCrCb);

Mat channel[3];
Mat result[3];

split(ycrcb,channel);

    Mat noise = Mat(channel[0].size(),CV_64F);


    normalize(channel[0], result[0], 0.0, 1.0, CV_MINMAX, CV_64F);
    randn(noise, 0, .1);
    result[0] = result[0] + noise;
    normalize(result[0], result[0], 0.0, 1.0, CV_MINMAX, CV_64F);

    result[0].convertTo(channel[0], CV_32F, 255, 0);

    imwrite("gaussiannoisy.jpg",channel[0]);

Mat output;
merge(channel,3,output);

imwrite("merge.jpg",output);

I can split the Y Cr, and Cb. I can add noise to Y too. But while I am trying to merge I am getting assertion error.

I have tried to separate the colour channels and add noise to each of them and merge them. And succeed to do so. But I want to separate the Y Cr Cb components and then add noise to only Y. Finally merge the components but could not. My code for colour channels:

for (int i=0;i<3;i++) {

    Mat noise = Mat(channel[i].size(),CV_64F);

    normalize(channel[i], result[i], 0.0, 1.0, CV_MINMAX, CV_64F);
    randn(noise, 0, .2);
    result[i] = result[i] + noise;
    normalize(result[i], result[i], 0.0, 1.0, CV_MINMAX, CV_64F);

    result[i].convertTo(result[i], CV_32F, 255, 0);
        }


merge(result,3,input);
imwrite("/Users/hossainmdshakhawat/developer/subimg/Noisy/merge.jpg",input);

I want to add gaussian noise to colour image where the standard deviation of gaussian noise were varied from 0.2 to 2 at 0.2 intervals.

Would please explain the line center = image(centerRect);

Actually I have done this using for loop and taking image an mat. But the problem is when accessing (x-1,y-1) point for x=0 and y=0. As it is out of matrix. Is there any way to test if a matrix position is out of range.

Thanks for you help. Would you please explain it more?

I want to apply a 3X3 window kernel in an image and replace the center pixel with the minimum difference between its surrounding pixels. Anybody please provide me with some help. I am new with OpenCV. And want to apply this process on whole image.