Background Question

Questions and answers about processing in StarTools and how to accomplish certain tasks.
Post Reply
gminder
Posts: 37
Joined: Sat Jul 27, 2013 2:58 am

Background Question

Post by gminder »

Ivo,

I am finding that I often get a mottled looking background in my images. The attached M27 image is an example that I finished tonight. Any advice on how I can smooth out the background to an even dark gray/black?

Gary
M27 Nebulosity Cal Reg Stack_ST Process_Reduced.jpg
M27 Nebulosity Cal Reg Stack_ST Process_Reduced.jpg (285.71 KiB) Viewed 5470 times
User avatar
admin
Site Admin
Posts: 3367
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Background Question

Post by admin »

Try increasing the largest scale (default at 0%) in that case (takes care of the chunkier noise 'blobs') in the Denoise module and perhaps lower the Scale correlation parameter (will ease off on trying to discover detail where none exists). Trying a Gaussian filter type instead might alos help. Noise reduction is always a tricky one - it's always a tradeoff!
Could you post the image as it looks before you apply noise reduction?
Ivo Jager
StarTools creator and astronomy enthusiast
gminder
Posts: 37
Joined: Sat Jul 27, 2013 2:58 am

Re: Background Question

Post by gminder »

I didn't save a pre-denoise version. I started over but the Life module did a great job this time and there wasn't anything left for the denoise module to do. The top photo was just before stopping tracking and the bottom photo was after de-noise.

For some reason I had a really hard time this time getting past dev-wipe with anything useable. At one point after doing an auto-dev with ROI I saw patches of red across the front of M27. But the image was so otherwise bad that I had to start over. I couldn't get back to where I could see the red detail. I am finding that in some of my images I have to use manual develop to avoid really bad distortion in the background.

I'm really happy with this latest attempt. I don't know if there is any additional red detail there but either way it's a nice image. https://www.dropbox.com/s/8fmh6rm9sd05u ... 0Stack.fit

Pre-Noise Reduction
M27 Nebulosity Cal Reg Stack_ST_Pre-Noise.jpg
M27 Nebulosity Cal Reg Stack_ST_Pre-Noise.jpg (262.34 KiB) Viewed 5456 times
Post Noise Reduction
M27 Nebulosity Cal Reg Stack_ST_Post-Noise_Reduced.jpg
M27 Nebulosity Cal Reg Stack_ST_Post-Noise_Reduced.jpg (135.07 KiB) Viewed 5456 times
User avatar
admin
Site Admin
Posts: 3367
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Background Question

Post by admin »

A very nice image indeed.
Can you tell us a little more about your acquisition setup and which programs you used to stack, etc?
I'm a bit puzzled as to why M27 is lacking red, while the stars exhibit all color ranges just fine in your final image! It's almost like the color channels (partiuclarly red) have been manipulated before being imported into StarTools. Another giveaway that not all channels are equal, is that the brighter (overexposed) stars are pink and not white. :think:

I'm thinking the 'red' you think you saw was possibly an illusion (color perception is an interesting psyhcological process ;) ). However, if you can replicate it, immediately launch the Color module and click 'Max RGB' this mode will show you which color channel is dominant (my bet is that it's not red!).

AutoDev will indeed cause problems with your data set, as it lacks flats and there is quite a bit of vignetting - AutoDev will keep bringing out this issue, unless Wipe can do a perfect job (which is *very* hard).
Ivo Jager
StarTools creator and astronomy enthusiast
Post Reply