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

DSLR Image Streaking

  • Please log in to reply
22 replies to this topic

#1 Aucello

Aucello

    Lift Off

  • *****
  • topic starter
  • Posts: 19
  • Joined: 04 May 2016

Posted 07 December 2021 - 07:10 PM

I am taking widefield mages using my modified Nikon D5300 with a 400 mm telephoto lens. I keep the exposures short (30 seconds each) but stack many (350 - 400) to get detail. I see horrible background noise in the form of streaking when I do this. Individual images fail to show this issue. I've attached a 3 hour image of the Horsehead to illustrate this. Any ideas on the source of thHorsehead 04 gradient_streaks.jpg is and how to process it out? GradientXTerminator fails to remove it.

 

Rick 



#2 Dynan

Dynan

    Fly Me to the Moon

  • *****
  • Posts: 6,120
  • Joined: 11 Mar 2018
  • Loc: NOLA

Posted 07 December 2021 - 07:12 PM

Looks like 'walking noise' from lack of dithering. Difficult or impossible to process out.

 

You won't see it in individual frames since it's an artifact of the stack.


Edited by Dynan, 07 December 2021 - 07:23 PM.


#3 nsblifer

nsblifer

    Explorer 1

  • -----
  • Posts: 72
  • Joined: 27 Oct 2021

Posted 07 December 2021 - 07:46 PM

Shouldn’t be that bad just because you didn’t dither. Assuming you used calibration frames?

#4 rickmurray1989

rickmurray1989

    Sputnik

  • -----
  • Posts: 43
  • Joined: 17 Jul 2017
  • Loc: Hendersonville, TN US

Posted 07 December 2021 - 07:54 PM

Pull it into Lightroom and lower the black levels. It will eliminate the streaks.



#5 17.5Dob

17.5Dob

    Cosmos

  • *****
  • Posts: 7,732
  • Joined: 21 Mar 2013
  • Loc: Colorado,USA

Posted 07 December 2021 - 08:03 PM

Walking noise......you might be tracking, but your mount  is also slowly drifting over the course of the entire stack. The drift is dragging out your stuck/hot pixels causing the streaks once you stack. You need to dither....


Edited by 17.5Dob, 07 December 2021 - 08:06 PM.

  • CharLakeAstro and brlasy1 like this

#6 nyairman

nyairman

    Messenger

  • -----
  • Posts: 449
  • Joined: 19 Aug 2017
  • Loc: new york

Posted 07 December 2021 - 08:42 PM

As the saying goes...dither or die. Star Tools has a tool to help with walking noise but it is always best to prevent it in the first place by dithering.

Edited by nyairman, 07 December 2021 - 08:42 PM.

  • Dynan and OldManSky like this

#7 Aucello

Aucello

    Lift Off

  • *****
  • topic starter
  • Posts: 19
  • Joined: 04 May 2016

Posted 07 December 2021 - 09:59 PM

I did use calibration frames, which also failed to disclose this issue. I like the idea about the "walking noise" effect, because I am not using an autoguider with these images, just mount tracking for each 30 second exposure. I've never included dithering in any of my imaging, so perhaps including autoguiding with dithering will minimize this.

 

Rick



#8 Aucello

Aucello

    Lift Off

  • *****
  • topic starter
  • Posts: 19
  • Joined: 04 May 2016

Posted 07 December 2021 - 10:02 PM

An issue with setting the black point to a higher value is that I lose a lot of nebulosity detail as a result. My preference is to preserve the nebulosity in a 3 hour composite image.

 

Rick


  • sharkmelley and Dynan like this

#9 17.5Dob

17.5Dob

    Cosmos

  • *****
  • Posts: 7,732
  • Joined: 21 Mar 2013
  • Loc: Colorado,USA

Posted 07 December 2021 - 11:30 PM

The problem you are going to face is there's no way of getting rid of it after the fact without a lot of compromises. Walking noise needs to be dealt with before it happens.

I rotated your jpeg image and used a Canon banding script 3X and adjusted the curves a bit. This is likely as best you'll be able to do.

Attached Thumbnails

  • CN-horsehead.jpg


#10 Ron359

Ron359

    Surveyor 1

  • *****
  • Posts: 1,608
  • Joined: 21 Apr 2008

Posted 07 December 2021 - 11:51 PM

I did use calibration frames, which also failed to disclose this issue. I like the idea about the "walking noise" effect, because I am not using an autoguider with these images, just mount tracking for each 30 second exposure. I've never included dithering in any of my imaging, so perhaps including autoguiding with dithering will minimize this.

 

Rick

I've never seen walking noise look like that or that bad...   If you've never dithered before and (assuming you've) gotten good images before, then how can it be walking noise?   If you used flats, it looks to me more like something went wrong with flats or perhaps a light leak in dark frames. Could also be something wrong with the camera.  Work through the possible problems in a series of tests.   Its easy enough to shoot plain dark frames in the house at various ISOs and to see if it shows up again. That will eliminate the possibility of sensor or dark frame problem.  Then test with flats of some dimly lit room shots.  If those are not the problem, then start shooting dark sky again with your 'normal' method, then dithering etc.   


  • nsblifer likes this

#11 sharkmelley

sharkmelley

    Fly Me to the Moon

  • *****
  • Posts: 5,232
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 08 December 2021 - 01:51 AM

You are seeing walking noise.  Unfortunately, with the Nikon D5300, calibration with darks is ineffective because of Nikon's raw data filtering:

https://www.cloudyni...ring/?p=9082246

 

Assuming you are already using rejection of outlier pixel values during stacking, the only solution is to use dithering during acquisition.

 

Mark


  • CharLakeAstro, Dynan and Borodog like this

#12 Borodog

Borodog

    Soyuz

  • *****
  • Posts: 3,771
  • Joined: 26 Oct 2020

Posted 08 December 2021 - 08:50 AM

This is exactly what you see when (1) your polar alignment is off and (2) your darks are not calibrating well either because (a) You haven't shot enough of them so they are inherently noisy themselves, (b) they do not match the temperature of the lights, or © what sharkmelley said. That dude knows his ****.



#13 galacticinsomnia

galacticinsomnia

    Surveyor 1

  • *****
  • Posts: 1,930
  • Joined: 14 Aug 2020
  • Loc: Pacific Northwest - Oregon

Posted 08 December 2021 - 07:54 PM

Yeah, walking noise.  You can get around it by dithering, and also by changing up your acquisition methodology, as well as post work.

Keeping in mind that the more you stretch the dark's to the lights, the more apparent your walking noise will be.
With that information, you should be able to make adjustments also to your post work, think blending and HDR techniques as well as selective stretching of your data with masks.

Dither or a lot of additional work in acquisition and post.

Good luck

Clear Skies !!



#14 Between17

Between17

    Mariner 2

  • -----
  • Posts: 221
  • Joined: 08 Apr 2019
  • Loc: Here on the wet coast of Canada

Posted 08 December 2021 - 08:02 PM

Here is a youtube video that shows a Pixinsight process to try and replace the walking noise with a more random background noise. I've never had super good luck with it but you may find it interesting.



#15 spasatelmaliboo

spasatelmaliboo

    Lift Off

  • -----
  • Posts: 9
  • Joined: 14 Jul 2021

Posted 09 December 2021 - 12:36 PM

Dont waste your time by trying to recover this image, make another one.

 

1. Maybe your polarscope needs to be calibrated. Check it in daytime.

2. Polar align well. Its means set up all your gears, frame the target (can be not accurate), balance, polar align, return to target.

3. If you cant dither or guide right now, here is a trick that works for me. If its possible, rotate your camera a little bit several times during the session.

4. Maybe check polar align after session, is it drifted?


  • Ron359 likes this

#16 nsblifer

nsblifer

    Explorer 1

  • -----
  • Posts: 72
  • Joined: 27 Oct 2021

Posted 27 December 2021 - 08:26 PM

I am taking widefield mages using my modified Nikon D5300 with a 400 mm telephoto lens. I keep the exposures short (30 seconds each) but stack many (350 - 400) to get detail. I see horrible background noise in the form of streaking when I do this. Individual images fail to show this issue. I've attached a 3 hour image of the Horsehead to illustrate this. Any ideas on the source of th Horsehead 04 gradient_streaks.jpgis and how to process it out? GradientXTerminator fails to remove it.

Rick


What stacking software are you using. I still don’t think this is from lack of dithering or guiding. Certainly possible you’re using incorrect settings for stacking, resulting in walking noise. I can set one parameter incorrectly in astropixelprocessor and create walking noise all over my image just like this.

#17 radon199

radon199

    Vostok 1

  • -----
  • Posts: 156
  • Joined: 29 Sep 2021
  • Loc: Vancouver, Canada

Posted 27 December 2021 - 09:09 PM

Just wanted to add that if you are not dithering with software you can still take steps to avoid walking noise. Every 20 or so subs, offset your framing. It only needs to be 20-50 pixels in RA and Dec, but doing that every 20 or so subs will be enough to hide the walking noise. This is because the fixed pattern stays the same, but when the subs are Star aligned it will be offset every 20 subs in the stack. This is Anouilh for it to be mostly rejected by the stacking process. Dithering is the same thing but usually at a smaller offset and more frequently.

#18 vidrazor

vidrazor

    Surveyor 1

  • *****
  • Posts: 1,888
  • Joined: 31 Oct 2017
  • Loc: North Bergen, NJ

Posted 28 December 2021 - 02:33 AM

Considering the mount you have, why AREN'T you dithering? I took a stab at it in Photoshop applying iterative Replace Color processes on the independent channels (as a luminosity process). I figured it was going to be futile, but it was fun at least LOL. Iterative Replace Color per channel (and sometimes all channels) can work to a degree with mild walking noise, but this is just crazy.

Attached Thumbnails

  • death.jpg


#19 Ron359

Ron359

    Surveyor 1

  • *****
  • Posts: 1,608
  • Joined: 21 Apr 2008

Posted 28 December 2021 - 11:59 AM

OP doesn't mention how many darks or calibrations frames used. This seems to be an effect of too many short exp. lights and not enough darks.  

 

https://clarkvision....on-vs-no-darks/


  • nsblifer likes this

#20 Basestar

Basestar

    Sputnik

  • -----
  • Posts: 31
  • Joined: 25 Jun 2013

Posted 30 December 2021 - 12:27 PM

Would OP be able to salvage the data from this session if they went out and shot more of the same target on different nights and dithered in subsequent sessions?


  • CharLakeAstro likes this

#21 vidrazor

vidrazor

    Surveyor 1

  • *****
  • Posts: 1,888
  • Joined: 31 Oct 2017
  • Loc: North Bergen, NJ

Posted 30 December 2021 - 01:06 PM

Would OP be able to salvage the data from this session if they went out and shot more of the same target on different nights and dithered in subsequent sessions?

If you shoot different nights the total stack would diffuse without manually dithering, because the different positions would accomplish the same. However you'd have to shoot short sets of about 50 subs over several nights and use about 50 from this session. Of course that's an impractical approach. If you manually dither in one night you should still only use a small subset of the existing subs. Best really to start fresh.



#22 Borodog

Borodog

    Soyuz

  • *****
  • Posts: 3,771
  • Joined: 26 Oct 2020

Posted 30 December 2021 - 05:48 PM

Try using Long Exposure Noise Reduction next time. You'll get half the number of subs per unit time, but they will be be as dark corrected as is possible.



#23 CharLakeAstro

CharLakeAstro

    Viking 1

  • *****
  • Posts: 705
  • Joined: 12 Jan 2015
  • Loc: Ontario

Posted 30 December 2021 - 09:31 PM

Yes exactly.... 

All is not lost. If OP can frame up the same shot, and collect a bunch more images of the same target, while dithering, can still integrate this raw data already collected into that stack

 

Would OP be able to salvage the data from this session if they went out and shot more of the same target on different nights and dithered in subsequent sessions?




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