[WM]: Geometrically invariant watermarking using feature points

[kanchana m <mani_kanchana@yahoo.co.in>]

 With respect to the paper 'Geometrically Invariant Watermarking Using Feature Points',P. Bas, JM. Chassery and B. Macq  (IEEE
Trans. Image Processing, vol. 11, no. 9, pp. 1014-1028.)  I have a small doubt. In the watermark detection phase, while calculating
the threshold depending on the probability of false alarm,  a parameter u which corresponds to a normalized gaussian law is used.
Can anybody please explain in detail on how to calculate 'u' or how it is derived.
 
Thanks in advance,
Kamala.


______________________________________________________________________________

Watermarking Mailing List - http://www.watermarkingworld.org/ml.html
To unsubscribe send email to "majordomo@watermarkingworld.org" with
"unsubscribe watermarking YOURMAIL" in the body.
______________________________________________________________________________