Re: [APML] Astronomical Image Enhancement Trial

[Aplanatic@aol.com]

Bert, Robert, Chris,

Thanks for your replies.

I wrote the software using Visual Basic and the algorithms are my own.  At the moment the human interface to my program is really poor.  When I improve it a bit I'll post a version of the program that you can play with.  Currently the program can only read .jpg and .bmp files, and can only output .bmp.  I'm no software whiz, so it may stay this way.  Reading .tif files looks like a real pain...  Bmp files are lossless and photoshop handles them, so all is not lost.  

Right now I'm trying various methods to improve the look of stars.  Suggestions for other image enhancement techniques are welcome.

Bert:  The algorithms were written specifically for color image enhancement, but work with monochrome images as well.  I'd love to try the algorithms on one of your images.  I can handle anything up to a couple of megabytes.  Send one along and I'll give it a try.  I've fooled around quite a bit with Tony's recently posted horsehead CCD image.  It's a wonderful image the way it is, of course, but the enhancements are interesting.

Our cabin in the Mojave is coming along quite well.  We now have a flush toilet.  What a luxury!

Robert:  I was very gentle on Richard's image.  The user can "turn up" the sharpening and noise reduction algorithms way beyond any sensible value and easily make a big mess of an image.

Chris:  The dark artifacts around the stars over the Lagoon were indeed enhanced by the sharpening algorithm.  They were present in the original image that I worked from.  On unprocessed images the algorithm does not produce these artifacts.  The sharpening technique is based more on deconvolution than USM, although it has elements of both (and some new stuff).

Dave Rowe

Bert wrote:
> Subtle but very significant improvements! The detail in M8 and the star
> definition are both considerably better. I don't know how much more
> improvement you can produce, but it looks mighty good right now. Have you
> tried any CCD images? I have tons of poor ones that could use considerable
> improvement! Let me know if you would like some and what image size is best
> for your modem speed. Or I can send a CD with lots on it.
> PS: How's that Mojave observatory coming?
> Bert

Robert wrote:
> I agree.  You gotta look, but there is enough difference between the two to
> make it well worthwhile to have done the extra processing.

> Are these programs you wrote or applications of existing programs?

Chris wrote:
> To me, it almost looks like a very mild USM was applied.  All of original
> nebulosity seems to be there but the image looks like the contrast has been
> boosted.  I also notice some slight dark artifacts around stars with the
> Lagoon Nebula.  Maybe this was because you applied it to an existing web res
> JPG image, as you mentioned?

> Interesting stuff!  I look forward to seeing more.


_______________________________________________
Astro-Photo mailing list
Astro-Photo@seds.org
http://seds.org/mailman/listinfo/astro-photo