Re: [APML] Some processing questions

[Alan Voetsch <alanv12952@yahoo.com>]

Hey Juan,


--- pleiades2004@pleiades-astrophoto.com wrote:
> even doing that you still have some dust or scratches, use a clone
> stamp
> tool, that you'll find in most imaging applications (e.g., Photoshop,
> Picture Window, Gimp). A clone stamp tool works by selecting adjacent
> pixels and copying them to replace the scratches and dust.

I've taken the day off from the computer. But you're right. I think
I'll find an image to practice on and see what develops.

> We're about to release a new version of PixInsight LE, including more
> tools
> and important improvements for existing ones. The documentation is
> being greatly improved, too.

I assume you'll post an update on this?

> > http://www.pbase.com/avoetsch12952/image/34172846
> Hey, that's a beautiful image! Two things you can easily improve are
> color
> balance and midtones. The image is somewhat purple to my eye. 

Oh yea, big time purple. Everything I tried didn't seem to help much.
I'd get it more red, or blue, but it didn't look right.

> there is
> more data from the galaxy that can be enhanced. The red histogram is
> clipped
> at the shadows, something that *always* must be avoided. This is a
> quick try:
> 
>
http://pleiades-astrophoto.com/tmp/pi-test-images/M31-avoetsch/34172846.M31Combine2copy-PI.jpg

Better, but to my eye it is too blue. I'll try again with the curves.
One problem I have is that I get too much light on my monitor during
the day, plus, I don't think this monitor is properly calibrated. The
last time I saw one of those 'step-bars', the last 4 steps on the left
were completely black. I'll look into that too.
 
> The threshold parameter of unsharp mask (USM) is used to apply USM
> selectively. Roughly explained, a copy of the original is blurred
> with a
> Gaussian filter, whose 'size' is given by the Radius parameter
> (standard
> deviation, actually) in Photoshop. Then the blurred copy is
> subtracted from
> the original. This reduces the importance of low-frequency components
> in the
> image (larger structures), and enhances higher frequency components
> (smaller
> structures), hence the sharpen effect. The threshold parameter is
> used in
> the following way. Each pixel of the blurred copy is subtracted from
> original pixels, but only for those pixels that differ more than the
> threshold value from their blurred counterparts. This avoids
> sharpening
> low-contrast areas, as the sky background, for example, where
> hopefully
> blurring doesn't change pixel values too much. However, this is a
> rough
> parameter that in general does not give really good results. Using
> masks to
> protect selected areas is much more efficient.

Wow. OK, I have a lot of work to do. I need to get Jerry's cd back into
the computer and pay attention.

Your input is helpful and appreciated.

Thanks,
Alan


		
_______________________________
Do you Yahoo!?
Declare Yourself - Register online to vote today!
http://vote.yahoo.com
_______________________________________________
Astro-Photo mailing list
Astro-Photo@seds.org
http://seds.org/mailman/listinfo/astro-photo