[WM]: Spatial Domain Image Watermarking

["christine" <christine@valorianconsulting.com>]

Hi there
 
I came across your paper titled "Spatial-Domain Image Watermarking Robust against Compression, Filtering, Cropping, and Scaling"
 
This paper has been a great help. But there are some parts which I don't understand. For Algorithm 3 (Visual components)
 
1. For i = 1 to n do:
a) Compute mi := maxj |xi . xj |/4, for all pixels j which are neighbors of pixel i on the image (there are up to eight neighbors);
mi can be regarded
as a kind of     discrete derivative at pixel i. To bound the value of mi
between 1 and 4, perform the following corrections:
i. If mi > 4 then mi := 4.
ii. If mi = 0 then mi := 1.
b) Compute the darkness of the i-th pixel as di := (70.xi).4/70 if xi < 70 and di := 0 otherwise. We consider a pixel to be dark if
its color level is below 70. The value of di lies between 0 and 4.
c) Compute the preliminary visual component of the i-th pixel as vi := max(mi, di).
 
2. For i = 1 to n compute the .nal visual component of the i-th pixel as Vi :=maxj vj , for all pixels j which are neighbors of i on
the image plus pixel i itself.
 
Does anybody understand why mi should be bounded between 1 - 4?
 
 
Thank you.
 
Best regards,
Christine D.
 



______________________________________________________________________________

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