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

Weird Stretch-related artifacts during processing

beginner
  • Please log in to reply
7 replies to this topic

#1 ThySheepKing

ThySheepKing

    Lift Off

  • -----
  • topic starter
  • Posts: 3
  • Joined: 15 Jan 2021
  • Loc: Porto, Portugal

Posted 20 January 2021 - 03:31 PM

Hello everybody,

 

This year marks the beginning of my journey with this hobby and after countless youtube videos it's pretty awesome to finally get to do it myself. 

 

In my last attempts of imaging Andromeda, which took place in two separate nights (1st time stacking data from multiple nights) I noticed a pretty big reduction in noise, which is to be expected with more data. I've also calibrated with some dark and bias frames, no flats though. My excitment was dialed down though, as after a few stretches in Photoshop I noticed that there is a pretty noticeable and sharp difference in noise between what gets stretched and, what I suspect, doesn't. This is also the first time I've noticed these (let's call it) "artifacts", probably because most of what I've tried to image so far (Orion and Pleiades) has been quite noisy all around. 

 

So, I've come to get some input on what might be the reason for this. Maybe some of you have experienced this before.

 

Thanks in advance!

 

Pedro

 

________________________

 

Equipment:

Camera - Canon EOS 2000D

Scope - RedCat 51 250mm

Mount - IOptron Skyguider Pro

No autoguiding

No filters 

 

Attached Thumbnails

  • Capture.JPG
  • Capture1.JPG

  • limeyx likes this

#2 fewayne

fewayne

    Apollo

  • *****
  • Posts: 1,195
  • Joined: 10 Sep 2017
  • Loc: Madison, WI, USA

Posted 20 January 2021 - 04:30 PM

Why would you suspect something isn't being stretched? Could you describe your process? I'm not at all sure what I'm looking at, here.



#3 ThySheepKing

ThySheepKing

    Lift Off

  • -----
  • topic starter
  • Posts: 3
  • Joined: 15 Jan 2021
  • Loc: Porto, Portugal

Posted 20 January 2021 - 05:07 PM

Why would you suspect something isn't being stretched? Could you describe your process? I'm not at all sure what I'm looking at, here.

Sorry if I wasn't clear or maybe I'm misusing the word "stretching". What I meant is, a "pretty noticeable and sharp difference in noise between..." what I'm trying to bring out (the galaxy) and the background space.  

 

My process is similar to what you'd see in Nebula Photos youtube tutorials. Generally some simple levels adjustments first, followed by a curves adjustment to better bring out the data, I remove background grandients if necessary (with dust & scratches filter method) and adjust saturation/vibrance a bit. I am still trying out things and getting adjusted with the processes.

 

To clear up, in the first picture you see a clear devide between a very noisy part of the galaxy and the less noisy background space, when what I was expecting to see was a more even noise level across the whole picture, but maybe I'm wrong in thinking this. 

 

Hopefuly it's more clear now, but let me know if there's more I can help with.

 

Thank you! 



#4 Midnight Dan

Midnight Dan

    Voyager 1

  • *****
  • Posts: 14,229
  • Joined: 23 Jan 2008
  • Loc: Hilton, NY, Yellow Zone (Bortle 4.5)

Posted 20 January 2021 - 05:55 PM

The artifacts I'm seeing looks like clipping of the black levels, and posterization of the darker areas.  My first thought is, are you capturing in jpeg rather than Raw?  Next, if you're processing in Photoshop, are you processing it as a 8-bit image rather than an 16-bit image?

 

-Dan


  • jdupton likes this

#5 jdupton

jdupton

    Gemini

  • *****
  • Posts: 3,097
  • Joined: 21 Nov 2010
  • Loc: Central Texas, USA

Posted 21 January 2021 - 10:49 AM

Pedro,

 

   I agree with Dan above. To me, it looks like some areas of the image were clipped to pure black somewhere in the processing. When you apply a stretch, the real data can be stretched and shows the natural noise of the image background but the clipped data is not stretched since you cannot stretch zero to anything but zero. When you adjust the levels later, the zero gets to be a very smooth uniform gray. It contains no noise because there was no noise to stretch later on in processing due to the earlier clipping.

 

   I would check out your initial raw frames. You state that you used both Bias and Dark frames. That may be the source of the clipping. I didn't notice what software you used to calibrate and preprocess with but you might try stacking the data again using only Dark Frame calibration.

 

   First, measure the image brightness in different areas of the raw frames. The minimum brightness of any place in the raw Light Frame should always be higher than the maximum brightness of the Dark Frame in the same area. When the Dark is brighter than the Light, clipping will happen.

 

 

John


Edited by jdupton, 21 January 2021 - 10:57 AM.


#6 fewayne

fewayne

    Apollo

  • *****
  • Posts: 1,195
  • Joined: 10 Sep 2017
  • Loc: Madison, WI, USA

Posted 21 January 2021 - 11:08 AM

I don't mean to be cranky but could you please briefly describe your calibration and integration process verbally? I'd rather not have to go watch the videos in order to help you; after all, I have my own process already :-).



#7 ThySheepKing

ThySheepKing

    Lift Off

  • -----
  • topic starter
  • Posts: 3
  • Joined: 15 Jan 2021
  • Loc: Porto, Portugal

Posted 21 January 2021 - 12:44 PM

Thank you for your suggestions so far.

 

I'll do some more experimenting with the stacking process to try to find the source of this problem. I've checked that I shot in RAW and that the image is 16bits, though it actually starts as a 32bit image which I then use Image->Mode to convert to 16bits using the Exposure and Gamma setting.

 

I am using DSS to stack my frames, with the recommended/default settings. I'm going to try John's recommendation of checking the brightness of the lights vs the darks and I'll do some more versions without calibration frames, etc. Also, I didn't know you could clip your photos in the stacking process to begin with. From what I've seen people tend to do that on the post-processing and I've been careful not to do that.

 

I'll post some feedback on my results once I'm done. Again, thanks for taking the time :)

 

Pedro



#8 Der_Pit

Der_Pit

    Vanguard

  • *****
  • Posts: 2,072
  • Joined: 07 Jul 2018
  • Loc: La Palma

Posted 21 January 2021 - 12:49 PM

I'll do some more experimenting with the stacking process to try to find the source of this problem. I've checked that I shot in RAW and that the image is 16bits, though it actually starts as a 32bit image which I then use Image->Mode to convert to 16bits using the Exposure and Gamma setting.

I don't know the software you use, but that somehow sounds scary.  Stacking should/has to happen in the linear representation of the data, so 'gamma' doesn't go in properly there?




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





Also tagged with one or more of these keywords: beginner



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics