How to remove the RGB noise from image?

I have some RGB images that contain many red, green and blue dots, How can I reomve them without affecting the quality, contrast or the colour of the image?

Is there any free code that I can use?

Here is an example

Thanks in advance

Answer 1

[Image Analyst]

Looks like salt and pepper noise. See my attached demos. They do a modified median filter.

[Muhamed Sewidan]

Thank you for your response, However, after applying your code the output and the input are the same, As the image contains RGB salt and pepper noise in the first place, therefore adding new noise then removing it was in vain.

I tried to apply the filter directly on the image but there is still too much noise.

[Image Analyst]

Tell me how you adapted it to change the thresholds in this line:

% Find the noise in the red.
noiseImage = (redChannel == 0 | redChannel == 255);

What did you use instead of 0 and 255? Attach your code.

Maybe you need to look at the pixels and find out what values the noise has. Maybe you should use 10 and 240, or something else.

[Image Analyst]

The attached code seems to run pretty well.

[Muhamed Sewidan]

Frankly, I kept everything untouched, except the adding noise part

[Muhamed Sewidan]

Here is how i edit it. Please correct me if it is wrong.

% Demo to add "salt and pepper" noise to a color image,
% then restore the image by removing this noise with a 
% modified median filter that acts only on the noise pixels
% and not upon "good" non-noise pixels.
clear all
clc
fontSize = 15;
addpath 'R:\Data\Fizeau\Nylon 6\1.537 & Conversion';
I = imread ('Parallel (3).bmp') ;
% Get the dimensions of the image.  numberOfColorBands should be = 3.
[rows, columns, numberOfColorBands] = size(I);
% Display the original color image.
subplot(3, 4, 1), imshow(I), title('Original Color Image','Font size', font size);
% Extract the individual red, green, and blue color channels.
redChannel = I(:, :, 1);
greenChannel = I(:, :, 2);
blueChannel = I(:, :, 3);
% Display the individual red, green, and blue color channels.
subplot(3, 4, 2), imshow(redChannel), title('Red Channel');
subplot(3, 4, 3), imshow(greenChannel), title('Green Channel');
subplot(3, 4, 4), imshow(blueChannel), title('Blue Channel');
% Median Filter the channels:
redMF = medfilt2(redChannel, [9 9]);
greenMF = medfilt2(greenChannel, [9 9]);
blueMF = medfilt2(blueChannel, [9 9]);
% Find the noise in the red.
noiseImage = (redChannel == 0 | redChannel == 255);
% Get rid of the noise in the red by replacing with median.
noiseFreeRed = redChannel;
noiseFreeRed(noiseImage) = redMF(noiseImage);

% Find the noise in the green.
noiseImage = (greenChannel == 0 | greenChannel == 255);
% Get rid of the noise in the green by replacing with median.
noiseFreeGreen = greenChannel;
noiseFreeGreen(noiseImage) = greenMF(noiseImage);

% Find the noise in the blue.
noiseImage = (blueChannel == 0 | blueChannel == 255);
% Get rid of the noise in the blue by replacing with median.
noiseFreeBlue = blueChannel;
noiseFreeBlue(noiseImage) = blueMF(noiseImage);

% Display the restored individual color channel images.
subplot(3, 4, 10), imshow(noiseFreeRed), title('Restored R');
subplot(3, 4, 11), imshow(noiseFreeGreen), title('Restored G');
subplot(3, 4, 12), imshow(noiseFreeBlue), title('Restored B');

% Reconstruct the noise free RGB image
rgbFixed = cat(3, noiseFreeRed, noiseFreeGreen, noiseFreeBlue);
subplot(3, 4, 9), imshow(rgbFixed), title('Restored Image');

[Image Analyst]

OK. Now did you try the latest code I attached? It seems to work reasonably well.

[Muhamed Sewidan]

Thanks it worked. I appreciate your help

However in the three lines of the histogram of the free green, red and blue, it shows me error in them.

[Image Analyst]

You can comment those out. I just used those to determine the thresholds.Thanks for Accepting the answer.

[Muhamed Sewidan]

Yes I comment them out. No, thanks to you for your help.

[nathiya v]

But when use medfilt2 it is degrading the image quality,so what filter can be used to remove the rgb dotted noise