Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Need image feedback - M33 with ASI533mc

  • Please log in to reply
9 replies to this topic

#1 DRK73

DRK73

    Viking 1

  • *****
  • topic starter
  • Posts: 759
  • Joined: 08 Mar 2013
  • Loc: Maryland, United States

Posted 25 October 2020 - 02:31 PM

Main problem is noise in the image. I'm trying to calibrate using dark flats. I've also tried using traditional darks and bias on a couple of other images, but using just dark flats worked all right on this image (which is far from perfect, but I think a successful "first" outing with the camera). 

 

There are some known issues I've got to deal with - mainly that these images are from the same stack of 74 2-min exposures (gain 100, offset 10, temp -10C) taken with AT72EDII at f/6, Bortle 6 skies using an L-pro filter. I know I probably need twice as much integration time. I'm dithering while capturing, and in post tried two different methods to try to handle the noise. 

 

Image 1 (see below) - dithering only, used multi scale linear transform before stretching, TGVDenoise after, and it's *okay*, but still noisier than I expected. 

 

 

 

 Image 2 (see below) - dithering, and I used drizzle integration in post as well as same denoising routines as in the first. Here to my surprise I get the appearance of "hot" pixels (bright blue, red, green dots) that I thought this method would eliminate. 

 

 

 

Also - have tried a couple emission/reflection nebula and...well. Images are noisier than I can handle in post. 

Is this really just a question of integration time, and if so, how much *should* I be shooting for? I wasn't necessarily expecting a free lunch, but these are approaching DSLR levels of noise. So any advice regarding settings, integration/exposure time, etc., would be greatly appreciated!



#2 DRK73

DRK73

    Viking 1

  • *****
  • topic starter
  • Posts: 759
  • Joined: 08 Mar 2013
  • Loc: Maryland, United States

Posted 25 October 2020 - 02:31 PM

Image #1 

 

 

Attached Thumbnails

  • M33 no drizzle.jpg

  • pyrasanth, Dynan and Mike in Rancho like this

#3 DRK73

DRK73

    Viking 1

  • *****
  • topic starter
  • Posts: 759
  • Joined: 08 Mar 2013
  • Loc: Maryland, United States

Posted 25 October 2020 - 02:32 PM

Image #2

 

 



#4 wrnchhead

wrnchhead

    Mercury-Atlas

  • *****
  • Posts: 2,600
  • Joined: 28 Aug 2017
  • Loc: NE Kansas

Posted 25 October 2020 - 02:38 PM

https://jonrista.com...duction-part-1/

Here’s a great write up on noise in astrophotography. But in short, yes more integration time equals less noise and you are still on the upward curve before diminishing returns kick in. Twice as much would probably be good. My personal experience is that 6 hours is the beginning of an acceptable noise level with my setup.

All that said, your image is pretty good! It’s not THAT noisy.
  • dswtan and DRK73 like this

#5 DRK73

DRK73

    Viking 1

  • *****
  • topic starter
  • Posts: 759
  • Joined: 08 Mar 2013
  • Loc: Maryland, United States

Posted 25 October 2020 - 02:43 PM

https://jonrista.com...duction-part-1/

Here’s a great write up on noise in astrophotography. But in short, yes more integration time equals less noise and you are still on the upward curve before diminishing returns kick in. Twice as much would probably be good. My personal experience is that 6 hours is the beginning of an acceptable noise level with my setup.

All that said, your image is pretty good! It’s not THAT noisy.

Thanks - having trouble uploading second image. The appearance of hot pixels in the drizzled image just really surprised me. 

 

I was also trying to get a start on the emission/reflection nebula IC348, and *that* was just a mess that I decided to skip trying to keep processing until I get more time on it. 



#6 Dynan

Dynan

    Aurora

  • *****
  • Posts: 4,578
  • Joined: 11 Mar 2018
  • Loc: NOLA

Posted 25 October 2020 - 02:57 PM

Nice image! Gotta love the 533!


  • wrnchhead likes this

#7 pyrasanth

pyrasanth

    Gemini

  • *****
  • Posts: 3,357
  • Joined: 08 Jan 2016

Posted 25 October 2020 - 03:09 PM

I have always found that OSC cameras are noiser than their mono siblings. They require a lot of extra integration time perhaps as many as twice as many subs over their mono counterparts. I always find under my light polluted skies (I'm bortle 8) gradients that can be difficult to deal with. I can only suggest taking many more subs. I generally don't use my QHY168C unless the sky is unusually dak and with fast optics like the RASA. Many short subs on the OSC seem to work better than less but longer subs.

 

If you were shooting mono I would suggest you try demure denoise which when used correctly can make a substantial noise reduction without damaging fine detail however it does not work for OSC images.


Edited by pyrasanth, 25 October 2020 - 03:12 PM.

  • DRK73 likes this

#8 Stelios

Stelios

    Voyager 1

  • *****
  • Moderators
  • Posts: 10,448
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 25 October 2020 - 03:59 PM

You eliminate hot pixels by using darks. You mention just using "dark flats." Dark flats are used to calibrate *FLATS*. Not lights. You need both darks (same gain, temperature and exposure as your lights) and dark flats (same gain and exposure as your flats). 

 

The calibration sequence is: Integrate darks into MD, integrate dark flats into MDF. Calibrate flats with MDF. Integrate flats into MF. Calibrate lights with MD and MF (and MB--bias--if available). Cosmetic correct calibrated lights using MD. Debayer. Star align. Integrate star aligned lights. 


  • dswtan, DRK73 and Dynan like this

#9 DRK73

DRK73

    Viking 1

  • *****
  • topic starter
  • Posts: 759
  • Joined: 08 Mar 2013
  • Loc: Maryland, United States

Posted 25 October 2020 - 10:48 PM

You eliminate hot pixels by using darks. You mention just using "dark flats." Dark flats are used to calibrate *FLATS*. Not lights. You need both darks (same gain, temperature and exposure as your lights) and dark flats (same gain and exposure as your flats). 

 

The calibration sequence is: Integrate darks into MD, integrate dark flats into MDF. Calibrate flats with MDF. Integrate flats into MF. Calibrate lights with MD and MF (and MB--bias--if available). Cosmetic correct calibrated lights using MD. Debayer. Star align. Integrate star aligned lights. 

 

I understand what you're saying and I've taken darks, but to be honest I'm just kind of aping what a few other folks are doing with their 533s as I'm on the low end of the learning curve with this camera (I just got it), so am trying out a variety of processing methods. But I am definitely aware that the most important factor is that I'm still just messing around with ~2 hours total integration time. 



#10 DRK73

DRK73

    Viking 1

  • *****
  • topic starter
  • Posts: 759
  • Joined: 08 Mar 2013
  • Loc: Maryland, United States

Posted 25 October 2020 - 10:49 PM

I have always found that OSC cameras are noiser than their mono siblings. They require a lot of extra integration time perhaps as many as twice as many subs over their mono counterparts. I always find under my light polluted skies (I'm bortle 8) gradients that can be difficult to deal with. I can only suggest taking many more subs. I generally don't use my QHY168C unless the sky is unusually dak and with fast optics like the RASA. Many short subs on the OSC seem to work better than less but longer subs.

 

If you were shooting mono I would suggest you try demure denoise which when used correctly can make a substantial noise reduction without damaging fine detail however it does not work for OSC images.

Haven't tried that - I've used multiscale linear with linear images then TGVdenoise after stretching, and have been happy with that for mono images. OSC is definitely a new thing for me.




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics