Does image functions support double value greater than 1.0 ?

Basically I have image containing pixel that can exceed 1.0. Take for example this 2x2 images:

       I = [ 0.1124   0.2354
             1.3244   1.2114  ];

As you can see two of the pixels are greater than 1.

Now for academic purpouse I need to blur and recover this image. Something like this: http://www.mathworks.it/it/help/images/ref/deconvwnr.html

Using functions like:

   fspecial
   imfilter
   deconvwnr, deconvlucy
   (and others...)

Could I got any wrong values? (I have tested it and it seems the functions are working normally...)

Please don't say that I should normalize to 0...1 because I know I should do it, I just want to know if the functions I mentioned could have problems with those values greater than 1

Answer 1

[Image Analyst]

Try it and see. Usually when MATLAB IPT functions accept floating point images, they must be in the range 0-1 and I think they often explicitly say that. This function doesn't say so explicitly but I'd guess that was the case. Just try it and see. Do the demo with the 0-1 image, and then multiply by 2 and try again and see if it looks weird.

[Gianfry]

I did some test and it seems that function like imnoise automatically cut values greater than 1.0 Example:

imnoise(I, 'gaussian', 0, 0.001);