Image of the day

Captured by
RICHARD COFER

Lunar Occultation of Jupiter (Egress)

My Account

New to Astromart?

Register an account...

Need Help?

Processing question: HELP!!!!

Started by Astrovale, 02/07/2008 09:29PM
Posted 02/07/2008 09:29PM | Edited 02/07/2008 11:21PM Opening Post
Hello guys
a question for the most experienced:

I have a large number of 600sec st11K images. I cannot process them all at once (my pc, although powerful, cannot cope): what would be the best way to go about it?
If I process smaller groups, stack them (via averaging) and then process the resulting averages as if they were single images stacking them also via averaging, would this work as if I were processing the whole batch and average-stack them?

I have an ongoing project that I cannot complete unless I find out how to solve this problem sad

thanks for any advice guys smile

Luca

www.amun-ra.org
Solar Imaging Website
Posted 02/08/2008 12:00AM #1
Hi Gianluca,
That will work fine. Just do so many at a time and then when you have stacked them all, stack the combined files as if they were just regular FITS.
I have done this when I had too many.

[SIZE="Large"][/SIZE][COLOR="Blue"][/COLOR] Floyd Blue grin
Amateur Imager
Posted 02/08/2008 03:03AM #2
Hi Gianluca,

You might also try summing your individual subs and then averaging the stacks together.

Try both and see which you prefer!

Jason
Posted 02/08/2008 03:16PM | Edited 02/08/2008 03:17PM #3
Luca,
You didn't say what processing program that you are using, so I can't give specific advice, but here are some guidelines:
-----Your averaging method that you described will only work well if the total of each group of subexposures is equal, otherwise some groups will have a greater contribution they should and some will have lesser contributions.
-----If you intend to use combined images to create a master combine, then each of those sub-combines should be sums (preferably into 32-bit).
----Here is a good procedure:
***Process each group of exposures this way: by normalizing, doing data rejection, and finally doing a sum combine.
***Combine the processed subexposure sums with a simple mean (if the groups are equal) or a sum combine. You don't want to do normalization or data rejection for this combine.

I hope this helps,

dan