Some problem in computation of matrix operations?

I have a matrix of 1024*966 for this matrix I want to compute mean and want to scale after this ? what is meaning of this and want to improve rsult by doing this but how can I? Final result will be nearly equal to matrix of 1024*966.

[Azzi Abdelmalek]

you said: "What is meaning of this" What does that mean?

[Walter Roberson]

Are you trying to "center and rescale" for the purposes of improving the condition number ?

Answer 1

[Image Analyst]

You can do

meanValue = mean(yourMatrix(:));
scaledMeanValue = scaleFactor * meanValue;

The meaning of this is so obvious that I don't know what to say other than it takes the mean and scales it - not sure why you even asked. I don't really know why you say you want to improve upon it. Of course I did it in two lines but if you never need the original mean value every, you can do it all in one line - that's about the only improvement I can possibly think of.

Of course the final result will be nothing like the original matrix and in fact is not even a matrix at all - it's a scalar, so I don't know why you assert that. If you do want a matrix where you want the mean inside a sliding window you can use conv2():

smoothedImage = conv2(double(yourMatrix), ones(3)/9, 'same');

but again, it will not be nearly equal to your original matrix. So, bottom line, I'm having a really hard time trying to understand what you're asking.