comp.lang.idl-pvwave archive
Messages from Usenet group comp.lang.idl-pvwave, compiled by Paulo Penteado

Home » Public Forums » archive » display many color images
Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend 
Return to the default flat view Create a new topic Submit Reply
Re: display many color images [message #8035 is a reply to message #7924] Thu, 30 January 1997 00:00 Go to previous message
sterner is currently offline  sterner
Messages: 106
Registered: February 1991
Senior Member
dwu@larry.cc.emory.edu (Ding Wu) writes:

> Anyone knows how to display many color images simultaneously.

> Splitting the color table will badly damage the color, and almost
> impossible for many, say 20, images.

> I am thinking writing something using fixed the color table, using 256
> colors to approximate all the colors, then images pixels looking for its
> most close representations in the color table. It defenitely changes the
> colors of the image, just fools human's eyes.

> Better idea, or any written codes, please.

Split each image into its RGB components, concatenate them as desired
into merged RGB components, and put them back together in 8 bits using
color_quan.

For example, two images IMG1 and IMG2 with color tables R1,G1,B1 and
R2, G2, B2 may be split into their RGB components:
RR1=R1(IMG1) & GG1=G1(IMG1) & BB1=B1(IMG1)
RR2=R2(IMG2) & GG2=G2(IMG2) & BB2=B2(IMG2)
(The color tables are obtained when you read the image or if it's
on the screen use tvlct,R1,G1,B1,/get).

Put these components together somehow. If they are the same size
one way is to put them side by side:
RR=[RR1,RR2] & GG=[GG1,GG2] & BB=[BB1,BB2]

Now convert back to 8 bits:
C = color_quan(RR,GG,BB,R,G,B)

To display:
tv,C
tvlct,R,G,B

color_quan does have it's limitations. You can play around with
several keywords but sometimes it just doesn't give good results.
Another trick is to save the new RGB components as the RGB parts
of a JPEG image and then read it back. I have found this to work
better in most cases. It would be nice to have a more direct access
to the JPEG color algorithm, maybe as an option to color_quan.

You might not expect this to work too well but I have found that it
actually does a pretty good job, surprised me when I first tried it.

Ray Sterner sterner@tesla.jhuapl.edu
The Johns Hopkins University North latitude 39.16 degrees.
Applied Physics Laboratory West longitude 76.90 degrees.
Laurel, MD 20723-6099
WWW Home page: http://fermi.jhuapl.edu/s1r/people/res/res.html
[Message index]
 
Read Message
Read Message
Previous Topic: Colour Postscript and 'colorbar'.
Next Topic: Re: Angstrom symbol?

-=] Back to Top [=-
[ Syndicate this forum (XML) ] [ RSS ] [ PDF ]

Current Time: Mon Oct 27 00:46:20 PDT 2025

Total time taken to generate the page: 0.02858 seconds