How to reverse the normalization function normc/normr?

I have used the function of "normr/normr" to normalize a matrix, but how can I reverse the normalization? Thank you very much!

[Image Analyst]

What happened to the original matrix? Why did you lose it?

I don't have the normr() function. Please list which toolbox it's in in the Products section below.

[kokin]

I didn't discarding the original data. I just want to know the way that how to reverse the normalized data using normc/normr.

Answer 1

[Matt J]

If you haven't discarded the original data, there is nothing to restore. The "restored" matrix is the original one that you already have. If you plan to throw that data away, the best thing would be to save the row-wise (or column-wise) norms and use them when needed to undo normalization, e.g.,

    A =

         1     5     9    13
         2     6    10    14
         3     7    11    15
         4     8    12    16

    >> nr=sqrt(sum(A.^2,2));  %row-wise norms

    >> B=bsxfun(@rdivide,A,nr)   %same as B=normr(A)

    B =

        0.0602    0.3010    0.5417    0.7825
        0.1091    0.3273    0.5455    0.7638
        0.1493    0.3483    0.5473    0.7463
        0.1826    0.3651    0.5477    0.7303

    >> Arev=bsxfun(@times,B,nr)  %reverse

    Arev =

        1.0000    5.0000    9.0000   13.0000
        2.0000    6.0000   10.0000   14.0000
        3.0000    7.0000   11.0000   15.0000
        4.0000    8.0000   12.0000   16.0000