Page 1 of 1

IC1805 and IC1848 with stock DSLR: feedback on workflow and postprocessing

Posted: Thu Oct 15, 2020 9:33 am
by polslinux
Hello all,
a couple of nights ago I finally had a crystal clear night after weeks of clouds, so I decided to pull an almost-all-nighter and image something from my balcony.
The targeted DSOs were the Heart and Soul nebulae. Here's the hardware I've used:
  • Nikon D7500 stock
  • Tamron 150-600 G2
  • iOptron SkyGuider Pro
and here's some data:
  • Bortle 5/6 location (SQM 19.63)
  • 177 lights @ 170mm, ISO 100, f/5, each 1m45s long
  • 20 darks taken after the lights
  • 75 flats taken the next morning using a lightbox (this one in particular: https://www.amazon.com/NXDRS-Tracing-Ut ... B07D57BQM6)
  • 75 dark flats taken after the flats
Here you can find some files, if you want to have a look: https://drive.google.com/drive/folders/ ... sp=sharing
  • final_ST.jpg is the file just out of StarTools
  • final_ST_GIMP.jpg is the file sharpened and hue-saturated using GIMP
  • stacked_dss.FTS is the stacked file using DSS as per the workflow on startools.org
  • stacked_astap.fits is the stacked file using ASTAP as per the workflow on startools.org
  • a single raw file for each type of frame (light, dark, flat and flat dark)

ST workflow:

Code: Select all

autodev -> keep
bin -> 60% -> keep
crop -> keep
wipe -> do -> keep
autodev -> redo global stretch -> keep
contrast -> dark anomaly filters 3 -> do -> keep
Sharp -> next -> keep
Color -> dark saturation 3.10, saturation 240%, green bias reduce 1.98 -> keep
Life -> less=More -> do -> keep
Entropy -> Dark/Light 70%/30% -> do -> keep
Tracking off -> grain removal: grain size 2 pixels -> next
I know that for those target I should have used a modified DSLR, but I'm not gonna buy one because it's very likely I will buy an astro-dedicated camera in a not-so-far future.
I know also that I should have taken at least 3 times more data, but that was the only single clear night I had in a while (and probably the only one for still a while) :cry:

Now some questions for you:
  1. are my calibration frames well taken? Or did I do something wrong with darks/flats/dark flats?
  2. why do I have so many green pixel in my final result? Is it because I need way more signal?
  3. why do I have this dense stars bloat in my final image? Is it because of the lens/too few signal/etc? I see online pictures with way less stars, how is it achievable without using software? Or is it just post-processing stuff?
  4. can I add extra data next time I will have a clear night? If yes, is it gonna be a problem if I have to re-focus because in the meanwhile I will have used my lens for other stuff? Getting the same f-ratio and focal length is not a problem, but same focus is never gonna be possible.
  5. with the data I have, how is my final result? Can it be improved? If yes, how?
  6. any other feedback for me?
Thanks a lot guys! :thumbsup:

Re: IC1805 and IC1848 with stock DSLR: feedback on workflow and postprocessing

Posted: Thu Oct 15, 2020 2:02 pm
by happy-kat
What version of DSS and startools are you using?
Did you take any bias files?

Re: IC1805 and IC1848 with stock DSLR: feedback on workflow and postprocessing

Posted: Thu Oct 15, 2020 2:20 pm
by polslinux
happy-kat wrote: Thu Oct 15, 2020 2:02 pm What version of DSS and startools are you using?
Did you take any bias files?
DSS 4.2.5 and ST 1.6.400MR1.
No, I didn't take any bias because I've read that they're not really useful if flats and dark flats are well taken.

Re: IC1805 and IC1848 with stock DSLR: feedback on workflow and postprocessing

Posted: Thu Oct 15, 2020 5:10 pm
by happy-kat
With a DSLR darks can be found to add noise, but bias are ok and help with DSLR. Your camera mileage may vary.
Your green is natural. The bayer matrix has 2 green pixels, 1 red, 1 blue.
If you are ensuring no white balance in DSS settings then in startools in the colour module select the matric for your camera.
Hope that helps

Re: IC1805 and IC1848 with stock DSLR: feedback on workflow and postprocessing

Posted: Fri Oct 16, 2020 2:38 am
by admin
Some quick observations;
  • The ASTAP stack appears to have been color balanced and there appears to be a dust donut remaining.
  • The DSS stack does not appear to have been color balanced and would make for a better starting point.
  • Your calibration frames do not appear to catch all the uneven lighting (and one notable smudge/dust speck remains at 9 o'clock).
  • Your lens suffers from chromatic aberration. You will want to use less aggressive saturation for the highlights, use the Repair Highlights parameter in the Color module and/or use the Filter module's Fringe Killer algorithm to get rid of purple fringing.
  • Overall, I think you did a nice job with the dataset at hand!

Re: IC1805 and IC1848 with stock DSLR: feedback on workflow and postprocessing

Posted: Fri Oct 16, 2020 7:53 am
by polslinux
admin wrote: Fri Oct 16, 2020 2:38 am Some quick observations;
  • The ASTAP stack appears to have been color balanced and there appears to be a dust donut remaining.
Hmm, that's weird. I followed exactly what I did the other time.
I have tried again, this time by modifying a setting. Could you please check whether this ASTAP stack is good?

Same link as the first post, but there's a new file called stacked_astap_NEW.fits

Also, how do you determine if a stack has been color corrected? Could you tell me please? So next time I'll be able to evaluate it myself :mrgreen:

Re: IC1805 and IC1848 with stock DSLR: feedback on workflow and postprocessing

Posted: Sat Oct 17, 2020 2:32 am
by admin
polslinux wrote: Fri Oct 16, 2020 7:53 am Hmm, that's weird. I followed exactly what I did the other time.
I have tried again, this time by modifying a setting. Could you please check whether this ASTAP stack is good?
Yep, that's better! :thumbsup:
Also, how do you determine if a stack has been color corrected? Could you tell me please? So next time I'll be able to evaluate it myself :mrgreen:
The bias color (when opened with the first option) was red, which is a tell-tale sign of balance correction. A non-whitebalanced dataset from a DSLR or OSC will never have a red bias, but almost always a green, teal or (in rarer cases) yellow bias.

Re: IC1805 and IC1848 with stock DSLR: feedback on workflow and postprocessing

Posted: Sun Oct 18, 2020 6:55 am
by polslinux
admin wrote: Sat Oct 17, 2020 2:32 am
polslinux wrote: Fri Oct 16, 2020 7:53 am Hmm, that's weird. I followed exactly what I did the other time.
I have tried again, this time by modifying a setting. Could you please check whether this ASTAP stack is good?
Yep, that's better! :thumbsup:
[/quote]

If you want to update the how-to on startools.org, this is what I did extra compared to last time: I disabled "Colour Smooth":

Re: IC1805 and IC1848 with stock DSLR: feedback on workflow and postprocessing

Posted: Mon Oct 19, 2020 6:02 am
by admin
Much appreciated! I updated the section about ASTAP. Thank you!