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

Colored Ring in Stacked Image

astrophotography dslr imaging optics Orion reflector beginner
  • Please log in to reply
52 replies to this topic

#1 Kugelblitz101

Kugelblitz101

    Sputnik

  • -----
  • topic starter
  • Posts: 36
  • Joined: 14 Jan 2018

Posted 25 May 2018 - 11:03 AM

Hello Cloudy Nights,

 

While processing an image of M13 taken with an unmodified Nikon D5500 I came across this apparent blue "ring" that appeared when I was stretching the image. This ring is apparent in my other images as well, so I believe it's caused by something in my optical setup rather than a specific target. It seems to become more apparent when the image is stretched and the saturation is increased. Here are the images I was looking at:

 

M13 Normal Image:

https://drive.google...rS-WSQGEWzQTpYV

 

M13 Saturation Increase:

https://drive.google...KVkGJ8vJBe4aSnE

 

Strangely, the ring is confined to the blue channel:

 

M13 Saturation Increase Blue Channel:

https://drive.google...IZeVruqGo4aaGNd

 

I found this thread that describes a similar issue which suggested the problem may be with the flat frames, so I stacked the image without flats and got these. I've applied the same stretch as the original image:

 

M13 No Flats Normal Image:

https://drive.google...OTAUX0c3LdVamhj

 

M13 No Flats Saturation Increase:

https://drive.google...Dx-NbxHmG1JzQn4

 

This time, the red and blue channels exhibit some form of a ring:

 

Red Channel:

https://drive.google...XufB03e1S9oSbBa

 

Blue Channel:

https://drive.google...VJcMLzn2DtCR2C1

 

All of the images were stacked with DeepSkyStacker. I took 30 flat frames by covering the objective of my Orion 8" Newtonian Astrograph with a t-shirt, placing a computer monitor with a blank notepad window in front of it, and exposing for 50% on back-of-camera histogram. I also took 30 dark flat frames. Stretching the master flat frame revealed an apparent banding:

 

Master Flat:

https://drive.google...p6K2FzngYzHwMMj

 

Stretched Master Flat:

https://drive.google...t-buyE42uTa2Aw_

 

Stretched Master Flat Histogram:

https://drive.google...Qnkz0At6W68MVji

 

Any advice or insight into this problem would be greatly appreciated! If any additional information is needed, just let me know!



#2 apophisOAS

apophisOAS

    Mariner 2

  • -----
  • Posts: 236
  • Joined: 20 May 2017

Posted 25 May 2018 - 12:19 PM

i never use The T-Shirt flat system , I rather use a metal ring with a pure white sheet of paper (without watermarks)and place it over the end of scope at first light during day adjusting iso and exposure time . Never had banding issues as a monitor is not as uniform as it may appear.

Roger



#3 TxStars

TxStars

    Mercury-Atlas

  • *****
  • Posts: 2,570
  • Joined: 01 Oct 2005
  • Loc: Lost In Space

Posted 25 May 2018 - 01:05 PM

Looks like you have a fall off in brightness, likely due to a smallish secondary or some other source of vignetting ..

Can you show your imaging train



#4 sharkmelley

sharkmelley

    Aurora

  • *****
  • Posts: 4,621
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 25 May 2018 - 01:45 PM

If you look at the back-of-camera histogram for your lights and flats, how far across is the peak of the histogram?

 

The might be due to underexposed flats.

 

On the other hand Jerry experienced a similar issue, this time using a D5300:  https://www.cloudyni...ipping-related/

I don't think that thread ever reached a conclusive diagnosis.

 

Mark



#5 Kugelblitz101

Kugelblitz101

    Sputnik

  • -----
  • topic starter
  • Posts: 36
  • Joined: 14 Jan 2018

Posted 25 May 2018 - 02:11 PM

Thank you all for your replies!

 

 

i never use The T-Shirt flat system , I rather use a metal ring with a pure white sheet of paper (without watermarks)and place it over the end of scope at first light during day adjusting iso and exposure time . Never had banding issues as a monitor is not as uniform as it may appear.

Roger

Unfortunately I don't have paper big enough to cover the entire objective, but I took flats with the t-shirt over the scope when it was pointed at the blue sky. I'll try them when DeepSkyStacker finishes stacking!

 

 

Looks like you have a fall off in brightness, likely due to a smallish secondary or some other source of vignetting ..

Can you show your imaging train

I attached the pictures to this post! The imaging train consists of the Baader MPCC Mark III, a Nikon T ring adapter, and the D5500. Let me know if you need any more information!

 

 

If you look at the back-of-camera histogram for your lights and flats, how far across is the peak of the histogram?

 

The might be due to underexposed flats.

 

On the other hand Jerry experienced a similar issue, this time using a D5300:  https://www.cloudyni...ipping-related/

I don't think that thread ever reached a conclusive diagnosis.

 

Mark

I'll take a look at that thread! The lights were shot at ISO 100 for 1 minute and peaked at 25%. The flats were shot at the same ISO and peaked at 50%. DeepSkyStacker is stacking an image of M101 with the same specifications for the light frames but with the flats peaking at 75% instead of 50%. I'll let you know if that changes anything as soon as it finishes!

Attached Thumbnails

  • Train 1.jpg
  • Train 2.jpg
  • Train 3.jpg


#6 TxStars

TxStars

    Mercury-Atlas

  • *****
  • Posts: 2,570
  • Joined: 01 Oct 2005
  • Loc: Lost In Space

Posted 25 May 2018 - 03:03 PM

After looking at the instruction manual for the Baader MPCC Mark III, I would say it is caused bu the Baader MPCC Mark III it's self.

https://www.baader-p...tion_manual.pdf

 

The images they use show the same thing.


Edited by TxStars, 25 May 2018 - 03:04 PM.


#7 gcardona

gcardona

    Ranger 4

  • -----
  • Posts: 380
  • Joined: 24 Nov 2009

Posted 25 May 2018 - 03:11 PM

Your flats are saturated. You can tell by the yellow tinge of the center area in the stretched image of your master flat, and also by the histogram. The right toe needs to be away from the edge, like around the 75% mark or your flats are over-exposed and in the non-linear response region of the sensor. You should try shorter exposures.



#8 Kugelblitz101

Kugelblitz101

    Sputnik

  • -----
  • topic starter
  • Posts: 36
  • Joined: 14 Jan 2018

Posted 25 May 2018 - 10:01 PM

Thank you again for your replies!

 

I stacked the M101 image with the flats taken with the monitor peaking at 75%. After doing a (very) quick stretch, the banding has apparently gotten worse in the blue channel and spread to the red channel. No saturation boost was added to these images.

 

Stretched Image:

Pinwheel Galaxy Stretch.jpg

 

Red Channel:

Pinwheel Galaxy Red Channel.jpg

 

Blue Channel:

Pinwheel Galaxy Blue Channel.jpg

 

After looking at the instruction manual for the Baader MPCC Mark III, I would say it is caused bu the Baader MPCC Mark III it's self.

https://www.baader-p...tion_manual.pdf

 

The images they use show the same thing.

Is there any way that this can be fixed with different flats or in post processing? Am I going to need Pixinsight?

 

Your flats are saturated. You can tell by the yellow tinge of the center area in the stretched image of your master flat, and also by the histogram. The right toe needs to be away from the edge, like around the 75% mark or your flats are over-exposed and in the non-linear response region of the sensor. You should try shorter exposures.

Should I aim for the peak to be in the 25% range?



#9 gcardona

gcardona

    Ranger 4

  • -----
  • Posts: 380
  • Joined: 24 Nov 2009

Posted 25 May 2018 - 10:40 PM

Thank you again for your replies!

 

 

Is there any way that this can be fixed with different flats or in post processing? Am I going to need Pixinsight?

 

Should I aim for the peak to be in the 25% range?

If you can take your flats again with a similar setup, you can try shorter exposures such that the histogram stays away from the right edge. Vignetting such as this can be take care of in post processing but it is difficult and there will usually be some residual gradients that will be hard to remove. It is much better to get your flats to work correctly.



#10 Kugelblitz101

Kugelblitz101

    Sputnik

  • -----
  • topic starter
  • Posts: 36
  • Joined: 14 Jan 2018

Posted 26 May 2018 - 03:40 PM

If you can take your flats again with a similar setup, you can try shorter exposures such that the histogram stays away from the right edge. Vignetting such as this can be take care of in post processing but it is difficult and there will usually be some residual gradients that will be hard to remove. It is much better to get your flats to work correctly.

Alright thank you! I'm away for the weekend so I don't have access to my scope, but I'll try this as soon as I get back!

 

In the meantime, I've been reading this thread that Mark posted and trying the solutions offered there. One of the replies suggestions was to disable "Set the black point to 0" and enable "Optimize Darks" in DSS. This seemed to fix a lot of the color issues in my images, such as this one of M13. It also eliminated the banding in the red and green color channels (they only have a slight gradient which I believe is from the sky as I'm in a white zone). For all of these tests, I stacked 20 lights, 10 darks, 10 bias, 10 flats, and 10 dark flats:

M13 Changed DSS Settings.jpg

 

But it seems to have done little to fix the blue ring. I also did a comparison on the M101 data between flats taken with the peak at 75% and those with the peak at 50%, both with the new DSS settings. These are the blue channels which still show the ring:

75% Flats with new DSS settings:

Pinwheel Blue Channel 75% Flats New DSS Settings.jpg

 

50% Flats with new DSS settings:

Pinwheel Blue Channel 50% Flats New DSS Settings.jpg

 

Is this ring an issue of vignetting or something totally different (such as a reflection or something in the MPCC)?



#11 AgilityGuy

AgilityGuy

    Viking 1

  • -----
  • Posts: 809
  • Joined: 20 Jun 2015
  • Loc: Northern CA

Posted 26 May 2018 - 03:58 PM

There are a number of Nikon users that have seen similar ringing patterns in our images.  Here was a recent thread - https://www.cloudyni...amyang-135mm/  



#12 entilza

entilza

    Soyuz

  • *****
  • Posts: 3,798
  • Joined: 06 Oct 2014
  • Loc: Oakville, ON, Canada

Posted 26 May 2018 - 07:14 PM

Shoot at ISO 400 next time give it a try.

#13 Kugelblitz101

Kugelblitz101

    Sputnik

  • -----
  • topic starter
  • Posts: 36
  • Joined: 14 Jan 2018

Posted 26 May 2018 - 09:47 PM

Shoot at ISO 400 next time give it a try.

Just flats or lights and flats?



#14 entilza

entilza

    Soyuz

  • *****
  • Posts: 3,798
  • Joined: 06 Oct 2014
  • Loc: Oakville, ON, Canada

Posted 26 May 2018 - 10:07 PM

Both. All new calibrations.

#15 17.5Dob

17.5Dob

    Fly Me to the Moon

  • ****-
  • Posts: 7,084
  • Joined: 21 Mar 2013
  • Loc: Colorado,USA

Posted 26 May 2018 - 10:12 PM

I think this has to be something very image chain specific, or a quirk in stacking, because I use the essentially identical D5300 and have never had a "ringing" issue shooting at ISO 200 to ISO 800 and f ratio from f2.8 to f7 in the past 3 years no matter the parameters..


Edited by 17.5Dob, 26 May 2018 - 10:18 PM.


#16 entilza

entilza

    Soyuz

  • *****
  • Posts: 3,798
  • Joined: 06 Oct 2014
  • Loc: Oakville, ON, Canada

Posted 26 May 2018 - 10:46 PM

I've reproduced this problem with 100 iso and a fast lens. Just with flats alone by dividing the separate channels by red for example. You can try the other dividing by the other channels. Mark helped narrow this down.

If this is the same issue a higher iso will help. The ISO 100 would make it more apparent.

If its a reflection well some flat paint may help. The easiest test would be on the next attempt try a bit higher ISO and see if it improves.

#17 AgilityGuy

AgilityGuy

    Viking 1

  • -----
  • Posts: 809
  • Joined: 20 Jun 2015
  • Loc: Northern CA

Posted 26 May 2018 - 11:31 PM

There are a number of reasons I've moved to using ISO800 full time with my D800E but since I've locked on this setting I haven't had to deal with the ringing in my images.  I've recently changed my camera/telescope connection which allows the light pollution filter to be placed at the level of the T ring, directly in front of the camera instead of immediately in back of the rear cell of the telescope - a difference of about 9 cm.  At some point I'd like to do the experiment of decreasing to ISO400 and see if the change makes a difference.  Given that the weather hasn't allowed me to do any imaging around a new moon since January I'm unlikely to do anything other than use the current setup.    



#18 Kugelblitz101

Kugelblitz101

    Sputnik

  • -----
  • topic starter
  • Posts: 36
  • Joined: 14 Jan 2018

Posted 27 May 2018 - 12:41 PM

Both. All new calibrations.

Alright thanks! I'll give it a shot!

 

 

I think this has to be something very image chain specific, or a quirk in stacking, because I use the essentially identical D5300 and have never had a "ringing" issue shooting at ISO 200 to ISO 800 and f ratio from f2.8 to f7 in the past 3 years no matter the parameters..

Thank you for the info! I'll try taking shots with and without the MPCC and with a larger T-ring to see if it helps or changes anything.

 

 

I've reproduced this problem with 100 iso and a fast lens. Just with flats alone by dividing the separate channels by red for example. You can try the other dividing by the other channels. Mark helped narrow this down.

If this is the same issue a higher iso will help. The ISO 100 would make it more apparent.

If its a reflection well some flat paint may help. The easiest test would be on the next attempt try a bit higher ISO and see if it improves.

Exactly what do you mean by dividing the channels? I'm not familiar with that procedure or what it does. I'll give increasing the ISO a shot! 



#19 Kugelblitz101

Kugelblitz101

    Sputnik

  • -----
  • topic starter
  • Posts: 36
  • Joined: 14 Jan 2018

Posted 27 May 2018 - 12:42 PM

There are a number of reasons I've moved to using ISO800 full time with my D800E but since I've locked on this setting I haven't had to deal with the ringing in my images.  I've recently changed my camera/telescope connection which allows the light pollution filter to be placed at the level of the T ring, directly in front of the camera instead of immediately in back of the rear cell of the telescope - a difference of about 9 cm.  At some point I'd like to do the experiment of decreasing to ISO400 and see if the change makes a difference.  Given that the weather hasn't allowed me to do any imaging around a new moon since January I'm unlikely to do anything other than use the current setup.    

Well best of luck! Thank you for all of your help and advice!



#20 skyler

skyler

    Surveyor 1

  • *****
  • Posts: 1,841
  • Joined: 16 Aug 2006
  • Loc: TGPNW

Posted 27 May 2018 - 02:35 PM

Hello,

 

 

There a number of threads on this issue including one I started on a D750. Many of the same players are involved in the discussion but there is no absolute answer to this particular occurrence of this artifact or condition.  I was hoping you could fill us in if you feel like you found a full solution to this issue or if you are simply doing your best to live with this and process it out as best as you can.  When you feel you have found a workable solution, please let us know.

 

Here is our discussion on this so far and it includes Jerry's D5300  as well.

https://www.cloudyni...k-modedunmoded/

 

Any thoughts would be appreciated.

 

 

Skylar.


Edited by skyler, 27 May 2018 - 02:37 PM.


#21 Kugelblitz101

Kugelblitz101

    Sputnik

  • -----
  • topic starter
  • Posts: 36
  • Joined: 14 Jan 2018

Posted 27 May 2018 - 09:21 PM

Hello,

 

 

There a number of threads on this issue including one I started on a D750. Many of the same players are involved in the discussion but there is no absolute answer to this particular occurrence of this artifact or condition.  I was hoping you could fill us in if you feel like you found a full solution to this issue or if you are simply doing your best to live with this and process it out as best as you can.  When you feel you have found a workable solution, please let us know.

 

Here is our discussion on this so far and it includes Jerry's D5300  as well.

https://www.cloudyni...k-modedunmoded/

 

Any thoughts would be appreciated.

 

 

Skylar.

Of course! I'll let you know as soon as I learn anything!



#22 sharkmelley

sharkmelley

    Aurora

  • *****
  • Posts: 4,621
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 28 May 2018 - 02:38 AM

In various threads I have seen, for those who have found a solution, the solution was increasing the ISO.  This is consistent with the way the internal data is treated i.e. digital scaling evidenced by the regular histogram gaps.  The Sony A7S is one of the worst offenders for this digital scaling and increasing the ISO works well.  We know that Nikons apply digital scaling in the red and blue channels but it is not nearly as pernicious.

 

However, not everyone has found a solution.  This makes me think that there are other issues that can lead to these rings.

 

Another cause of coloured rings on Sony cameras (using lenses) is the crude vignetting correction that is applied to the data.  If the camera supports the lens (i.e. the firmware recognises it and it has a table of corrections to apply) then it applies a correction to the raw data even when the corrections are disabled in the menu.  I would be surprised if Nikon is doing the same thing (i.e. applying corrections that are switched off) but it cannot be completely ruled out.  Try taping over the lens contacts (so the lens cannot be recognised) and see if it still happens.  The lens will need to be operated manually of course. 

 

For the avoidance of doubt, I should add that no camera applies vignetting corrections to telescopes because they cannot recognise the scope electronically.

 

Mark


Edited by sharkmelley, 28 May 2018 - 02:41 AM.


#23 whwang

whwang

    Soyuz

  • *****
  • Posts: 3,557
  • Joined: 20 Mar 2013

Posted 28 May 2018 - 02:41 AM

Interesting observations, Mark.

 

If the camera applies a vignetting correction, shouldn't it be equally applied to light and all the calibration frames?  Its effect should be flattened out once a full calibration is made.



#24 Kugelblitz101

Kugelblitz101

    Sputnik

  • -----
  • topic starter
  • Posts: 36
  • Joined: 14 Jan 2018

Posted 28 May 2018 - 07:21 PM

I went through some of my older data captured with the same camera/telescope setup and found that I imaged the Whirlpool Galaxy at ISO 800. A quick stack of 20 lights and 10 of each calibration frame revealed the same blue ring pattern, so I'm not sure that increasing the ISO is a solution to this particular problem.

 

Whirlpool Galaxy:

Whirlpool Blue Ring Color.jpg

 

Whirlpool Galaxy Blue Channel:

Whirlpool Blue Ring Blue Channel.jpg

 

I'm going to try changing a few other parameters in my setup (such as removing the MPCC) in order to help isolate the cause of this problem.



#25 ccs_hello

ccs_hello

    Voyager 1

  • *****
  • Posts: 10,719
  • Joined: 03 Jul 2004

Posted 28 May 2018 - 09:21 PM

One of the effect of applying flat is to correct vignetting (in camera term: lens shading) due to uneven image circle illumination.

We know it's not always perfect, especially when in-camera image processing (Expeed DSP) kicks in, in the following aspect: 

 

post-4145-14073976685916.jpg

 

(Here I am not talking about black clipping <-- it's from a D5100) but Red-channel and Blue-channel scaling.

The notches are due to math multiplication caused aliasing.

You can find such effect on first post's last image link (flat histogram of the D5300.)

 

Earlier suggestion on using high-enough ISO (say 800, when signal amplification is still analog amp) IMO was to get the histogram spread apart far enough thus you see many small gaps as opposed to using low ISO (say, 100) which may condense the histogram just in a small region, thus fewer notches but gaps are wider.

 

What Nikonhacker firmware mod <-- has several useful features for astro:

- disable Blue and Red channel scaling (i.e., no notches)

- change the black clipping point (N appears to have addressed in later cameras) so you get a bell shape

- make the low tier camera (such as D5100) be able to have (a) uncompressed NEF or (2) lossless NEF <-- note low tier only has lossy NEF (companding and Huffman code)

The net effect is to have this camera behaves as if it's almost an astroimager and can provide true sensor raw without Expeed doctoring it.

(Sorry no active cooling and still need to replace factory UV/IR-cut filter.)

 

Unfortunately, I don't think the extensive work done by Simeon and Plane were carried over to D5300 (due to it's new ARM based architecture, I read the forum to be that way...)

 

P.S.   D5100's image sensor is the popular IMX071, same one used is ASI071, QHY..., etc.

 

Clear Skies!

 

ccs_hello


Edited by ccs_hello, 28 May 2018 - 09:28 PM.

  • whwang likes this


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: astrophotography, dslr, imaging, optics, Orion, reflector, beginner



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics