Thanks Carlos, I will be working on it.
----- Original Message -----
Sent: Monday, January 31, 2005 2:08
AM
Subject: Re: [APML] Widefield Comet
Machholz
It isn't as complex as you may think. Each layer contains features of
certain characteristical scales, wich are very related to the pixel
scale that is shown. It depends critically with the scaling function you
use. If you are familiar with convolutions (user defined kernels, or maks),
you can think on the first layer just like a high-passed version of the
image, made with the same principles of the unsharp mask, but using the
scaling function as a bluring convolution. To create the other layers, you
make new high passed version of the image from the result of deleting all
the small scale features you just isolated in the first layer (so, now you
have a blured images as a remannent, called "residual image"), and so
on.
So, in practical words, we can think on every layer as a subimage
containing all the features of the original image at certain scales, plus a
last layer with the residual image. The final image is builded just by adding
each processed (smoothed, biased, etc.) layer to the residual image.
You just have to keep in mind the the way the features are separated into
those layers is in direct relation with the scaling function. If it looks very
sharp, you can isolate high frecuency features very well (for planetary
images, for example). In the other hand, big and plain functions will play a
great role if you want to enhance large scale features. Keep in mind that you
can manage the functions just like if they were convolution filters, modifying
them, creating new ones, etc.
Regards,
Carlos Milovic
F.
-------------------------
Astro & Photo -
CMF
http://www.astrophoto.vze.com
-------------------------
Visita
"AstroFoto", el foro de astrofotografía en
español
http://espanol.groups.yahoo.com/group/astrofoto
_______________________________________________
Astro-Photo
mailing
list
Astro-Photo@seds.org
http://seds.org/mailman/listinfo/astro-photo