[Numpy-discussion] meaning of accumulation/normalisation

wilson wilson.t.thompson at gmail.com
Sun Apr 13 04:30:22 EDT 2008


 the Frobenius norm,
> Which is matlab speak for sqrt(trace(dot(X.T, X)))

thanks for that one Chuck..
1.
i have seen similar normailzations in most image processing code..what
exactly is the purpose of such normalization?before making images
these values will have to be reprocessed to get the pixel values.being
a beginner this question has been bugging me for a while.if anyone can
explain it wd be great help
2.in this case,the java doc mentions that by the above steps ul is
normalised to unit
length (vector length) ,and that accumOperation(newul) should return
1,
i tested it
norm=sqrt(trace(dot(ul.T, ul)))
newul=ul/norm
testnorm=sqrt(trace(dot(newul.T, newul))) ====>1

can someone make clear the maths in it
thanks
W



More information about the NumPy-Discussion mailing list