Re: [WM]: Question on associate eigenvectors

[aweke.lemma@philips.com]

Dear deepthi,

you should look at the notations carefully. In my previous e-mail, I have assumed that e1x and e1y are the two eigen vectors of the
matrix M (i.e., column-vectors). In that case, the operation you just showed cannot be conducted (multiplying a 2x2 matrix with a
4x2 matrix). I would first understand the notations used in the paper and then try to understand the operations. (eg., are e1y and
e1x just entries or vector? are they row- or column-vectors? etc)

Regards,

Aweke. 
--------------------------------------------
Aweke Negash Lemma
Philips Digital System Labs
Eindhoven
http://www.pdsl.philips.com
---------------------------------------------
Tel. +31 40 2734092
Fax. +31 40 2737353









deepthi reddy <deep_269@yahoo.co.in>
Sent by: 
watermarking-owner@watermarkingworld.org
2004-10-18 02:57 AM
 
        To:     <watermarking@watermarkingworld.org>
watermarking-owner@watermarkingworld.org
        cc:     (bcc: Aweke Lemma/EHV/PDSL/PHILIPS)
        Subject:        Re: [WM]: Question on associate eigenvectors
        Classification: 




thank you for the reply. 
in that case e1x and eiy would be two column matrices. for example e1x=[a;b] but in the next step of processing i have :-
[a1 a2;a3 a4]= [c1 0;0 c2]*[e1x e1y;-e1y e1x] 
 
i dont know how i should do this.it would be great if you can clear this.
 
deepthi
 

e.lemma@philips.com wrote:
e1x and e1y should be the two columns of v. (In matlab code: e1x = v(:,1) and e1y = v(:,2)).

Regards,

--------------------------------------------
Aweke Negash Lemma
Philips Digital System Labs
Eindhoven
http://www.pdsl.philips.com
---------------------------------------------
Tel. +31 40 2734092
Fax. +31 40 2737353


deepthi reddy
Sent by: 
watermarking-owner@watermarkingworld.org
2004-10-14 03:51 PM

To: 
cc: (bcc: Aweke Lemma/EHV/PDSL/PHILIPS)
Subject: [WM]: Question on associate eigenvectors
Classification: 




Hello everyone,

Iam new to this field. Iam present working on a paper on robust digital watermarking, and i encountered this problem while
normalising the image.

there is a 2 by 2 matrix M. what the paper says exactly is " take its associate eigenvectors [e1x e1y] "

using the following command in matlab:-
[v d]=eig(M) , i got a 2 by 2 matrix for v. 
can someone tell me what [e1x e1y] would be?

deepthi


______________________________________________________________________________

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