Adding high density salt and pepper noise to an image

I am taking 'cameraman.tif' as my input image.

How can I add high density salt and pepper noise to it so that if I place 3X3 window over it and scan for uncorrupted pixels in it the uncorrupted pixels will come out to be 3?

Please provide matlab code for this

[Roaa Mohammed]

clc; img = imread('cameraman.tif'); na = imnoise(img,'salt & pepper',0.02); black=3; white=253; nb = img; Rmatrix = randint(size(img,1),size(img,2),[0,255]); nb(Rmatrix <= black) = 0; nb(Rmatrix >= white) = 255; subplot 131;imshow(img),title('ORIGINAL IMAGE'); subplot 132;imshow(na),title('NOISE USING FUNCTION'); subplot 133;imshow(nb),title('NOISE WITHOUT USING FUNCTION');

Answer 1

[Jaimin]

Hi @Avinash Bhatt

To add high-density salt-and-pepper noise to an image such that only 3 uncorrupted pixels remain in any 3x3 window, you will need to carefully select the noise density. The salt-and-pepper noise will randomly turn some pixels to either 0 (pepper) or 255 (salt).

Kindly refer following code snippet for better understanding.

img = imread('cameraman.tif');
img = im2double(img);

windowSize = 3;

% Calculate the total number of pixels in a 3x3 window
totalPixelsInWindow = windowSize^2;

% Define the number of uncorrupted pixels desired
uncorruptedPixels = 3;

% Noise density is the proportion of corrupted pixels
noiseDensity = (totalPixelsInWindow - uncorruptedPixels) / totalPixelsInWindow;

noisyImg = imnoise(img, 'salt & pepper', noiseDensity);

noisyImg = im2uint8(noisyImg);

For more information kindly refer to following MathWorks documentation.

imnoise: https://www.mathworks.com/help/images/ref/imnoise.html