How to add uniform noise to image with median and standard deviation

Hello..

what is the direct command for "uniform noise" with specific median and standard deviation, for example if the median =40 and the standard deviation=20 how can I write the noise command to add noise to the following code of original image:

I=imread('cameraman.tif');

original=imshow(I)

thank you

Answer 1

[Matt J]

I've never heard of "noise" with non-zero mean/median, but generating it is straightforward,

delta=standardDeviation*sqrt(12);

noiseTerm = (rand(size(I))-1/2)*delta + Median;

I = I + noiseTerm;

You can also use unifrnd() if you have the Statistics and Machine Learning Toolbox.

noiseTerm = unifrnd(Median-delta/2 , Median+delta/2, size(I));

[Mahabba Almheiri]

Can you clarify please:

1- What is delta meaning ?

2- From where you got 12 in sqrt(12) ?

thank you so much

[Matt J]

delta is the width of the uniform distribution, i.e., if the variable is distributed on the interval [a,b] then delta=b-a.

The sqrt(12) comes from the formula for the variance of a uniform random variable

variance=(b-a)^2/12

[William Rose]

@Mahabba Almheiri, @Matt J is right. IN case it is not obvious, what Matt is doing is scaling up the range foe the uniform noise so it will have the standard deviaiton you want it to have. Uniform noise that ranges from A to B (range=B-A) has a standard deviation

Like @Matt J, I don;t understand the use of "median" here. It is just adding a constant "brightness" to the whole image. Are you worried about negative values when you add noise?