WMW
  NEWS     ABOUT     PARTNERS     CONTACT  
  HELP
WATERMARKING
  • Biography
  • FAQ
MAILINGLIST
  • Management
  • Archive
CONFERENCES
  • Calls
BENCHMARKING
  • Stirmark
  • CheckMark
  • Optimark
BOOKS
LINKS
  • Companies
  • Research
  • Others
WEBRING
DISCLAIMER

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [WM]: MATLAB problem in color Image Processing



Dear tdhien

Thanks for your reply,. I truly appreciate it.
 but still there is a difference between c7 and c3 which are supposed to b= e the
same
 and also between moda and x which are also supposed to be the same
 
 do you have any suggestions why this difference
occur?
 
 Thanks a lot
 
 Dahlia
 
 
 -------Original Message-------
 
From: Thai, Hien Duy
Date: 02 يوليو, 2005 11:11:21 م
To: watermarking@watermarkingworld.org
Subject: Re: [WM]: MATLAB problem in color Image Processing
 
 
 Dear ahlia,
  
 You may make a mistake when convert the
data type between uint8 (0- 255,
8-bit) and double
Change: "save modlisa.mat atmon" intead of "save modlisa.mat catmon8"  in
the first program, nd   "mor=catmon" intead of "mor=catmon8"  in the  second
program.  You multiply all the coefficients by 100, and then the data is over "uint8". 
 Usually, when one watermark color image.
Color images are converted in to=  a YIQ representation and the rightness component Y is then watermarked. The color image can then
be converted to other formats but should be converte= d back to YIQ to extraction watermark. Or you can directly watermark into R= ,
G  and B respectively.
  
 tdhien
 
 xc=imread('monalisa.bmp');
 figure(1)
 imshow(xc)
 %Title('The Original Image
256 * 256')
 x1=xc(:,:,1);
 x1=double(x1);
 x2=xc(:,:,2);
 x2=double(x2);
 x3=xc(:,:,3);
 x3=double(x3);
x=x3;
[c3,s3]=wavedec2(x,2,'db1');     % size of ch2,cv2,cd2 is 4761*3=14283
c4=c3.*100;
s4=s3;
mod=waverec2(c4,s4,'db1');
catmon=cat(3,x1,x2,mod);
catmon8=uint8(catmon);
%save modlisa.mat catmon8
save modlisa.mat catmon;
figure(2)
imshow(catmon8)
imwrite(catmon8,'modlisa.bmp')

load modlisa.mat
%mor=imread('modlisa.bmp');
%mor=catmon8;
mor=catmon;
figure(1)
imshow(mor)
Title('The Original Image')
y1r=mor(:,:,1);
y1r=double(y1r);
y2g=mor(:,:,2);
y2g=double(y2g);
y3b=mor(:,:,3);
y3b=double(y3b);
y= yR>3b;
[c6,s6]=wavedec2(y,2,'db1');
c7=c6./100;
s7=s6;
moda=waverec2(c7,s7,'db1');
catmona=cat(3,y1r,y2g,moda);
catmona8=uint8(catmona);
figure(2)
imshow(catmona8)
image([catmona8])
imwrite(catmona8,'modlisab2n.bmp')



----- Original Message ----- 
From: Dahlia 
To: watermarking@watermarkingworld.org 
Sent: Friday, July 01, 2005 8:35 PM
Subject: [WM]: MATLAB problem in color Image Processing
 
 
 
 
 Hello
 
 I am working in the field of Digital Image
Watermarking for olored images= 
 
 Attached to this email are two programs
 In order to process colored images as in the 1st
program
 
 -I first separated the three RGB components from one nother
 -then I performed a DWT on the Bule image component
-and then I tried to multiply all the coefficients by a onstant number (1= 00)
 -Then performed an IDWT on the modified blue image
omponent
 -and Finally, I combined the RG and modified blue omponents back together and wrote the modified image
 
 In the 2nd
program
 
 I read the modified RGB image, separated the three omponents
 and then performed a DWT on the blue modified image
omponent
 Then I divided all the coefficients by the same constant used in the 1st stage (100)
 Then I performed an IDWT on the
modified blue image omponent
 and finall combined the RG and B all together and wrote the processed ima= ge
 
 matter of fact,
there is a difference between both images, the original Monalisa Image and the processed image.
 The precision have been doubled
for all variables.
 When the value of the constant change the difference etween the original = and processed images change
 when
the mul;tiplication and division are changed into ddition and subtraction, for values below 100, there is a slight modifications,
for values below 1000 , they are more, and so on
 
 so I want to ask why is the process not reversible?
 In other meanings
the subtraction should cancel the effect of addition
 and the division should cancel the effect of the multiplication,
 but this
is not what happens
 
 
 This problem doesn't appear in gray-level images; and a olored image is supposed to be a three stacks
of gray level images, so I really cannot explain why this happens and  I would truly appreciate a further assistan= ce to solve
it.
 
 
 Also, if you can direct me for further readings in the area of watermarki= ng colored images, I would truly
appreciate it. As a matter of fact, I could only find 3 papers on the internet and the three dds PN sequences while I would like to
add a picture which involves multiplying and dividing by th= e weighing factor (ALPHA)
 Thanks in Advance
===========================
Eng. Dahlia R. ElShafie
Teaching Assistant nd M.Sc. Student
Electronics and communications
College of ngineering and Technology
 Arab Academy for Science and Technology
 Alexandria-Egypt
 
 ________
 
Watermarking Mailing List - http://www.watermarkingworld.org/ml.html
To unsubscribe send email to "majordomo@watermarkingworld.org" with
 "unsubscribe watermarking YOURMAIL" in the body.
_____________________________________________________________________________
 






Internal Virus Database is out-of-date.
 Checked by AVG Anti-Virus.
Version: 7.0.323 / Virus Database: 267.6.2 - Release ate: 2005/06/04 --------------Boundary-00=_CQP0G6G0000000000000
Content-Type: Text/HTML;
  charset="windows-1256"
Content-Transfer-Encoding: quoted-printable

<meta http-equiv="Content-Type" content="text/html; charset=windows= -1256"><html> <head> <meta name="GENERATOR"
content="IncrediMail 1.0">
 <!--IncrdiXMLRemarkStart> <IncrdiX-Info> <X-FID>FLAVOR00-NONE-0000-0000-000000000000</X-FID>
<X-FVER></X-FVER>
<X-CNT>;</X-CNT>
</IncrdiX-Info>
<IncrdiXMLRemarkEnd-->

</head>

<BODY background="" bgColor=#ffffff style="BACKGROUND-POSITION: 0px=  0px; FONT-SIZE: 10pt; MARGIN: 1px; FONT-FAMILY: Arial"
scroll=yes ORGY= POS="0" X-FVER="2.0"> <TABLE border=0 cellPadding=0 cellSpacing=0 id=INCREDIMAINTABLE width="95%"> <TR>

<TD id=INCREDITEXTREGION width="100%" style="PADDING-RIGHT: 7px; PADDING-LEFT: 7px; FONT-SIZE: 10pt; FONT-FAMILY: Arial" 
   >
      <DIV>
      <TABLE id=INCREDIMAINTABLE cellSpacing=0 cellPadding=0 width== "95%" 
      border=0>
        
        <TR>
          <TD id=INCREDITEXTREGION 
          style="PADDING-RIGHT: 7px; PADDING-LEFT: 7px; FONT-SIZE: 10pt= ; FONT-FAMILY: Arial" 
          width="100%">
            <DIV>Dear <SPAN 
            style="FONT-SIZE: 10pt; FONT-FAMILY: rial">tdhien</SPAN></DIV>
            <DIV><SPAN 
            style="FONT-SIZE: 10pt; FONT-FAMILY: rial"></SPAN>&nbsp;</DIV>
            <DIV><SPAN style="FONT-SIZE: 10pt; FONT-FAMILY: rial">Thank= s for 
            your reply,. I truly appreciate it.</SPAN></DIV>
            <DIV><SPAN style="FONT-SIZE: 10pt; FONT-FAMILY: rial">but s= till 
            there is a difference between c7 and c3 which are supposed to=  be the 
            same</SPAN></DIV>
            <DIV><SPAN style="FONT-SIZE: 10pt; FONT-FAMILY: rial">and a= lso 
            between moda and x which are also supposed to be the 
            same</SPAN></DIV>
            <DIV><SPAN 
            style="FONT-SIZE: 10pt; FONT-FAMILY: rial"></SPAN>&nbsp;</DIV>
            <DIV>do you have any suggestions why this difference occur?</DIV>
            <DIV>&nbsp;</DIV>
            <DIV>Thanks a lot</DIV>
            <DIV>&nbsp;</DIV>
            <DIV>Dahlia</DIV></TD></TR></TABLE></DIV>
      <DIV>&nbsp;</DIV>
      <DIV id=IncrediOriginalMessage><I>-------Original Message-------<= /I></DIV>
      <DIV>&nbsp;</DIV>
      <DIV id=receivestrings>
      <DIV dir=ltr style="FONT-SIZE: 11pt" <i><B>From:</B></I> <A 
      href="mailto:tdhien@augusta.eee.u-ryukyu.ac.jp">Thai, Hien Duy</A></DIV>
      <DIV dir=ltr style="FONT-SIZE: 11pt" <i><B>Date:</B></I> 02 ي= وليو, 2005 
      11:11:21 م</DIV>
      <DIV dir=ltr style="FONT-SIZE: 11pt" <i><B>To:</B></I> <A 
      href="mailto:watermarking@watermarkingworld.org">watermarking@watermarkingworld.org</A></DIV>
      <DIV dir=ltr style="FONT-SIZE: 11pt" <i><B>Subject:</B></I> Re:=  [WM]: 
      MATLAB problem in color Image Processing</DIV></DIV>
      <DIV>&nbsp;</DIV>
      <DIV>&nbsp;</DIV>
      <DIV><SPAN style="FONT-SIZE: 10pt; FONT-FAMILY: Arial">
      <P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN 
      style="FONT-SIZE: 10pt; FONT-FAMILY: Arial">Dear 
      ahlia,<o:p></o:p></SPAN></P>
      <P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN 
      style="FONT-SIZE: 10pt; FONT-FAMILY: rial"><o:p>&nbsp;</o:p></SPAN></P>
      <P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN 
      style="FONT-SIZE: 10pt; FONT-FAMILY: Arial">You may&nbsp;make a m= istake 
      when convert the data type between uint8 (0- 255, 8-bit)&nbsp;and 
      double<o:p></o:p></SPAN></P>
      <P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN 
      style="FONT-SIZE: 10pt; FONT-FAMILY: Arial">Change: "save modlisa= .mat 
      atmon" intead of "save modlisa.mat catmon8"&nbsp; in the first prog= ram, 
      nd&nbsp; &nbsp;"mor=catmon"&nbsp;intead of "mor=catmon8"&nbsp; = in 
      the&nbsp; second program. <SPAN 
      style="mso-spacerun: yes">&nbsp;</SPAN>You&nbsp;multiply all the 
      coefficients by 100, and then the data is over "uint8". 
      <o:p></o:p></SPAN></P>
      <P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN 
      style="FONT-SIZE: 10pt; FONT-FAMILY: Arial">Usually, when one wat= ermark 
      color image. Color images are converted in to a YIQ representation = and the 
      rightness component Y is then watermarked. The color image can then=  be 
      converted to other formats but should be converted back to YIQ to 
      extraction watermark. Or you can directly watermark into R, G, and = B 
      respectively.<o:p></o:p></SPAN></P>
      <P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN 
      style="FONT-SIZE: 10pt; FONT-FAMILY: rial"><o:p>&nbsp;</o:p></SPAN></P>
      <P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><SPAN 
      style="FONT-SIZE: 10pt; FONT-FAMILY: rial">tdhien<o:p></o:p></SPAN></P>
      <P class=MsoNormal 
      style="MARGIN: 0cm 0cm 0pt"></SPAN><BR>xc=imread('monalisa.bmp');<BR>figure(1)<BR>imshow(xc)<BR>%Title('The 
      Original Image 256 * 
      256')<BR>x1=xc(:,:,1);<BR>x1=double(x1);<BR>x2=xc(:,:,2);<BR>x2=double(x2);<BR>x3=xc(:,:,3);<BR>x3=double(x3);<BR>x=x3;<BR>[c3,s3]=wavedec2(x,2,'db1');&nbsp;&nbsp;&nbsp;&nbsp; 
      % size of ch2,cv2,cd2 is 
      4761*3=14283<BR>c4=c3.*100;<BR>s4=s3;<BR>mod=waverec2(c4,s4,'db1');<BR>catmon=cat(3,x1,x2,mod);<BR>catmon8=uint8(catmon);<BR>%save 
      modlisa.mat catmon8<BR>save modlisa.mat 
      catmon;<BR>figure(2)<BR>imshow(catmon8)<BR>imwrite(catmon8,'modlisa.bmp')</P>
      <P class=MsoNormal style="MARGIN: 0cm 0cm 0pt"><BR>load 
      modlisa.mat<BR>%mor=imread('modlisa.bmp');<BR>%mor=catmon8;<BR>mor=catmon;<BR>figure(1)<BR>imshow(mor)<BR>Title('The 
      Original 
      Image')<BR>y1r=mor(:,:,1);<BR>y1r=double(y1r);<BR>y2g=mor(:,:,2);<BR>y2g=double(y2g);<BR>y3b=mor(:,:,3);<BR>y3b=double(y3b);<BR>y= 
      yR&gt;3b;<BR>[c6,s6]=wavedec2(y,2,'db1');<BR>c7=c6./100;<BR>s7=s6;<BR>moda=waverec2(c7,s7,'db1');<BR>catmona=cat(3,y1r,y2g,moda);<BR>catmona8=uint8(catmona);<BR>figure(2)<BR>imshow(catmona8)<BR>image([c
>atmona8])<BR>imwrite(catmona8,'modlisab2n.bmp')<BR></P></DIV>
      <DIV>&nbsp;</DIV>
      <DIV>&nbsp;</DIV>
      <BLOCKQUOTE 
      style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
        <DIV style="FONT: 10pt arial">----- Original Message ----- </DIV>
        <DIV 
        style="BACKGROUND: #e4e4e4; FONT: 10pt arial; font-color: lack"><B>From:</B> 
        <A title=dollissima@hotmail.com 
        href="mailto:dollissima@hotmail.com">Dahlia</A> </DIV>
        <DIV style="FONT: 10pt arial"><B>To:</B> <A 
        title=watermarking@watermarkingworld.org 
        href="mailto:watermarking@watermarkingworld.org">watermarking@watermarkingworld.org</A> 
        </DIV>
        <DIV style="FONT: 10pt arial"><B>Sent:</B> Friday, July 01, 2005 8:35 
        PM</DIV>
        <DIV style="FONT: 10pt arial"><B>Subject:</B> [WM]: MATLAB prob= lem in 
        color Image Processing</DIV>
        <DIV><BR></DIV>
        <TABLE id=rem_INCREDIMAINTABLE cellSpacing=0 cellPadding=0 = width="95%" 
        border=0>
          
          <TR>
            <TD id=rem_INCREDITEXTREGION 
            style="PADDING-RIGHT: 7px; PADDING-LEFT: 7px; FONT-SIZE: 10= pt; FONT-FAMILY: Arial" 
            width="100%">
              <DIV>&nbsp;</DIV>
              <DIV>&nbsp;</DIV>
              <DIV>Hello</DIV>
              <DIV>&nbsp;</DIV>
              <DIV>I am working in the field of Digital Image Watermarkin= g for 
              olored images</DIV>
              <DIV>&nbsp;</DIV>
              <DIV>Attached to this email are two programs</DIV>
              <DIV>In order to process colored images as in the 1st  program</DIV>
              <DIV>&nbsp;</DIV>
              <DIV>-I first separated the three RGB components from one 
              nother</DIV>
              <DIV>-then I performed a DWT on the Bule image component</DIV>
              <DIV>-and then I tried to multiply all the coefficients by = a 
              onstant number (100)</DIV>
              <DIV>-Then performed an IDWT on the modified blue image 
              omponent</DIV>
              <DIV>-and Finally, I combined the RG and modified blue ompo= nents 
              back together and wrote the modified image</DIV>
              <DIV>&nbsp;</DIV>
              <DIV>In the 2nd program</DIV>
              <DIV>&nbsp;</DIV>
              <DIV>I read the modified RGB image, separated the three 
              omponents</DIV>
              <DIV>and then performed a DWT on the blue modified image 
              omponent</DIV>
              <DIV>Then I divided all the coefficients by the same consta= nt 
              used&nbsp;in the&nbsp;1st stage&nbsp;(100)</DIV>
              <DIV>Then I performed an IDWT on the modified blue image 
              omponent</DIV>
              <DIV>and finall combined the RG and B all together and wrot= e the 
              processed image</DIV>
              <DIV>&nbsp;</DIV>
              <DIV>matter of fact, there is a difference between both ima= ges, 
              the original Monalisa Image and the processed image.</DIV>
              <DIV>The precision have been doubled for all variables.</DIV>
              <DIV>When the value of the constant change the difference e= tween 
              the original and processed images change</DIV>
              <DIV>when the mul;tiplication and division are changed into=  
              ddition and subtraction, for values below 100, there is a s= light 
              modifications, for values below 1000 , they are more, and s= o 
              on</DIV>
              <DIV>&nbsp;</DIV>
              <DIV>so I want to ask why is the process not reversible?</DIV>
              <DIV>In other meanings</DIV>
              <DIV>the subtraction should cancel the effect of addition</DIV>
              <DIV>and the division should cancel the effect of the 
              multiplication,</DIV>
              <DIV>but this is not what happens</DIV>
              <DIV>&nbsp;</DIV>
              <DIV>&nbsp;</DIV>
              <DIV>This problem doesn't appear in gray-level images; and = a 
              olored image is supposed to be a three stacks of gray level=  
              images, so&nbsp;I really cannot explain why this happens 
              and&nbsp;&nbsp;I would truly appreciate a further assistanc= e to 
              solve it.</DIV>
              <DIV>&nbsp;</DIV>
              <DIV>&nbsp;</DIV>
              <DIV>Also, if you can direct me for further readings in the=  area 
              of watermarking colored images, I would truly appreciate it= . As a 
              matter of fact, I could only find 3 papers on the internet = and the 
              three dds PN sequences while I would like to add a picture = which 
              involves multiplying and dividing by the weighing factor 
              (ALPHA)</DIV>
              <DIV>Thanks in Advance</DIV>
              <DIV>===========================</DIV>
              <DIV>
              <DIV><STRONG><FONT color=#804000 size=3>
              <DIV><FONT face="Times New Roman" size=3><STRONG><FONT = size=3 
              olor="#804000">
              <DIV><FONT face="Times New Roman" size=2>Eng. Dahlia R.=  
              ElShafie</FONT></DIV>
              <DIV><FONT face="Times New Roman" size=2>Teaching Assis= tant nd 
              M.Sc. Student</FONT></DIV>
              <DIV><FONT face="Times New Roman" size=2>Electronics an= d 
              communications</FONT></DIV>
              <DIV><FONT face="Times New Roman" size=2>College of = ngineering 
              and Technology</FONT></DIV>
              <DIV><FONT face="Times New Roman" size=2>Arab Academy f= or Science 
              and Technology</FONT></DIV>
              <DIV><FONT face="Times New Roman" 
              size=2>Alexandria-Egypt</FONT></DIV>
              <DIV></FONT></STRONG></FONT></FONT></STRONG>&nbsp;</DIV>
              <DIV>________<BR><BR>Watermarking Mailing List - <A 
              href="http://www.watermarkingworld.org/ml.html">http://www.watermarkingworld.org/ml.html</A><BR>To 
              unsubscribe send email to "<A 
              href="mailto:majordomo@watermarkingworld.org">majordomo@watermarkingworld.org</A>" 
              with<BR>"unsubscribe watermarking YOURMAIL" in the 
              body.<BR>______________________________________________________________________________ 
              </DIV></DIV></DIV></DIV></TD></TR>
          <TR>
            <TD id=rem_INCREDIFOOTER width="100%">
              <TABLE cellSpacing=0 cellPadding=0 width="100%">
                
                <TR>
                  <TD width="100%"></TD>
                  <TD id=rem_INCREDISOUND vAlign=bottom align=middle></TD>
                  <TD id=rem_INCREDIANIM vAlign=bottom 
                align=middle></TD></TR></TABLE></TD></TR></TABLE>
        <P>
        <HR>

        <P></P>Internal Virus Database is out-of-date.<BR>Checked by AVG 
        Anti-Virus.<BR>Version: 7.0.323 / Virus Database: 267.6.2 - Relea= se ate: 
        2005/06/04<BR></BLOCKQUOTE>
      <DIV>&nbsp;</DIV></TD>
</TR>

<TR>
<TD id=INCREDIFOOTER width="100%">

	<TABLE cellPadding=0 cellSpacing=0 width="100%"> 	<TR> 	<TD width="100%"></TD> 	<TD align=middle
id=INCREDISOUND vAlign=bottom></TD> 	<TD align=middle id=INCREDIANIM vAlign=bottom></TD> 	</TR> 	</TABLE>

</TD>
</TR>

</TABLE>
</BODY>
</html>

© 2000-2002 by WatermarkingWorld
Design and Concept by Martin Kutter