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 processing of globular clusters

DSLR Imaging
  • Please log in to reply
17 replies to this topic

#1 condensermike

condensermike

    Explorer 1

  • -----
  • topic starter
  • Posts: 83
  • Joined: 03 Nov 2012
  • Loc: Virginia

Posted 18 June 2021 - 11:56 AM

Hello folks,

 

I was wondering if anyone had some great resources for processing globular clusters? I did find a thread on here that referenced a now defunct Yahoo Groups page, which is a shame since it sounded like that person was an expert and even had sample TIFFs to work from. Does anything like this exist on this forum that I am overlooking? 

 

I am using a Canon 1100D / T3 and have it attached prime focus to a Meade R5 achromat. Here is a link to my astrobin page showing some of the results I have so far with fumbling around with processing. 
https://www.astrobin.../condensermike/

 

As you can see, one of my main issues is the horrible blue color bloat around stars thanks to the achromat lens in my scope. Any advice or links to good resources would be appreciated. 

 

Thanks in advance!

 

Mike


Edited by condensermike, 18 June 2021 - 11:57 AM.


#2 Jeffmar

Jeffmar

    Apollo

  • -----
  • Posts: 1,462
  • Joined: 18 Mar 2012
  • Loc: salt lake city, utah

Posted 18 June 2021 - 12:50 PM

I have seen filters available that can take out some of the blue halo around objects. I also think that most post processing apps have the ability to increase or decease he saturation of different colors in photos. I usually use Adobe Lightroom for post processing. It is simpler than Photoshop and does most of what I need to do.


Edited by Jeffmar, 18 June 2021 - 12:51 PM.

  • condensermike likes this

#3 galacticinsomnia

galacticinsomnia

    Apollo

  • *****
  • Posts: 1,459
  • Joined: 14 Aug 2020
  • Loc: Pacific Northwest - Oregon

Posted 18 June 2021 - 04:37 PM

On the bigger stars, I think you should be able to use either the Lens Correction and Color in Adobe Camera Raw, or in Affinity.

If you only have a few in each image Just make a selection of the star core, invert your selection, and you should be able to use the paintbrush or clone stamp, to remove the halo.
There are scripts available that can help, but often, I find them lacking the human interaction needed to remove the halos.

I have an Svbony scope that says its an Apochromatic element, but I still get halos, just not as bad as that in your current images there.

Hope you get it worked out.

Clear Skies !!


  • condensermike likes this

#4 the Elf

the Elf

    Soyuz

  • *****
  • Posts: 3,768
  • Joined: 06 Sep 2017
  • Loc: Germany

Posted 18 June 2021 - 05:29 PM

There is some talk going on here.

https://www.cloudyni...m92/?p=11164112

Freddy and me both use the Canon 800D / T7i. Hope that helps.

If you send a link to the unprocessed stack I can give it a go in PI.

Did you use deconvolution or any other sharpening technique?


Edited by the Elf, 18 June 2021 - 05:30 PM.

  • condensermike likes this

#5 KLWalsh

KLWalsh

    Apollo

  • -----
  • Posts: 1,077
  • Joined: 19 Feb 2014
  • Loc: North Georgia, USA

Posted 18 June 2021 - 07:42 PM

A pale yellow or amber filter, or a ‘fringe killer’ filter, will help remove the blue halo.
  • condensermike likes this

#6 BQ Octantis

BQ Octantis

    Fly Me to the Moon

  • *****
  • Posts: 5,198
  • Joined: 29 Apr 2017
  • Loc: Red Centre, Oz

Posted 18 June 2021 - 10:47 PM

G'day Mike,

 

The problem with an achro is that the blue is not in focus. If you look at your channels individually, you can see this very clearly:

 

(Click for full size.)

gallery_273658_12412_44566.jpg

 

Your peak focus is in the red channel (simply judging from its having the tightest stars and best contrast in the globular). There are several options to use that to your advantage. You can paste the red channel into the Luminance channel of a Lab version of your image, for instance. But the most effective method I've found is to paste the red channel as a new layer over the RGB image, and then set the new layer mode to Darken:

 

(Click for animated toggle with the original.)

gallery_273658_12412_295014.jpg

 

Hope that helps…

 

BQ


Edited by BQ Octantis, 18 June 2021 - 10:54 PM.

  • condensermike likes this

#7 condensermike

condensermike

    Explorer 1

  • -----
  • topic starter
  • Posts: 83
  • Joined: 03 Nov 2012
  • Loc: Virginia

Posted 19 June 2021 - 09:15 AM

BQ thanks for the tips and I plan to implement this soon. And thanks to everyone who has posted. 


Edited by condensermike, 19 June 2021 - 09:15 AM.


#8 condensermike

condensermike

    Explorer 1

  • -----
  • topic starter
  • Posts: 83
  • Joined: 03 Nov 2012
  • Loc: Virginia

Posted 19 June 2021 - 09:16 AM

G'day Mike,

 

The problem with an achro is that the blue is not in focus. If you look at your channels individually, you can see this very clearly:

 

(Click for full size.)

gallery_273658_12412_44566.jpg

 

Your peak focus is in the red channel (simply judging from its having the tightest stars and best contrast in the globular). There are several options to use that to your advantage. You can paste the red channel into the Luminance channel of a Lab version of your image, for instance. But the most effective method I've found is to paste the red channel as a new layer over the RGB image, and then set the new layer mode to Darken:

 

(Click for animated toggle with the original.)

gallery_273658_12412_295014.jpg

 

Hope that helps…

 

BQ

BQ, do you think the focus of my blue channel is particularly bad in the scope I am using or is this just inherent to achromat refractors in general?  



#9 RedLionNJ

RedLionNJ

    Aurora

  • *****
  • Posts: 4,610
  • Joined: 29 Dec 2009
  • Loc: Red Lion, NJ, USA

Posted 19 June 2021 - 11:10 AM

You have the common "not all colors come to the same focus" issue with any achromat. There are multiple ways to correct this in post-processing, or even prevent it from happening at all (some mentioned above).

 

Smaller stars are always a positive when displaying globular images. Deconvolution (even Registax works for this) is one of the better methods to reduce star size (given good data).

 

The other two techniques I would apply (totally up to you, of course) are being more gentle with the black-clipping (a really deep indigo background can be very pleasing, and also reveal more "outlier" stars) and a significantly heavier degree of saturation. A good saturation level will bring out the icy blues and warm yellows much more obviously and the contrast in these two colors can be simply beautiful in some globulars.  Under extreme situations, I've even converted my images to Lab color space and applied an S-curve to the relevant plane (I can never remember if it's a or b until I try it) to further separate the blues and yellows.

 

It appears you have good data (for an AR) - make the most of it.


  • condensermike likes this

#10 BQ Octantis

BQ Octantis

    Fly Me to the Moon

  • *****
  • Posts: 5,198
  • Joined: 29 Apr 2017
  • Loc: Red Centre, Oz

Posted 19 June 2021 - 05:55 PM

BQ, do you think the focus of my blue channel is particularly bad in the scope I am using or is this just inherent to achromat refractors in general?  

This is quite typical of achromats. Their lens design only brings two points across the color spectrum in focus on a single focal plane. This is generally red and green—blue is the odd man out. There is a nice plot of how far out of focus the visible spectrum is by lens type on wikipedia:

 

lenses.jpg

[Source; author = Cmglee]

 

There's no amount of deconvolution that will bring an achromat's 3rd channel into focus.


  • galacticinsomnia likes this

#11 galacticinsomnia

galacticinsomnia

    Apollo

  • *****
  • Posts: 1,459
  • Joined: 14 Aug 2020
  • Loc: Pacific Northwest - Oregon

Posted 19 June 2021 - 08:03 PM

This is quite typical of achromats. Their lens design only brings two points across the color spectrum in focus on a single focal plane. This is generally red and green—blue is the odd man out. There is a nice plot of how far out of focus the visible spectrum is by lens type on wikipedia:

 

attachicon.giflenses.jpg

[Source; author = Cmglee]

 

There's no amount of deconvolution that will bring an achromat's 3rd channel into focus.

That is informative.  Is it true that a good field flattener can also help, with bloat by their use of optical voodoo ?  

Clear Skies !!

 



#12 BQ Octantis

BQ Octantis

    Fly Me to the Moon

  • *****
  • Posts: 5,198
  • Joined: 29 Apr 2017
  • Loc: Red Centre, Oz

Posted 19 June 2021 - 08:20 PM

Is it true that a good field flattener can also help, with bloat by their use of optical voodoo ? 

I've not read this anywhere. Just filters.

 

Some time ago, I pondered capturing the RGB channels separately for my f/2.8 lens to correct its CA before I had an effective method to eliminate it. I've moved on since then (I use RawTherapee for the CA and an aperture mask to crisp up the stars), but if I had an achro I'd seriously consider this. In particular, I'd shoot the red and green separate from the blue. I just wonder if its focuser has enough range to bring the blue channel into focus.

 

BQ



#13 Thomas A Davis

Thomas A Davis

    Surveyor 1

  • *****
  • Posts: 1,708
  • Joined: 25 Mar 2005

Posted 19 June 2021 - 09:20 PM

I've not read this anywhere. Just filters.

 

Some time ago, I pondered capturing the RGB channels separately for my f/2.8 lens to correct its CA before I had an effective method to eliminate it. I've moved on since then (I use RawTherapee for the CA and an aperture mask to crisp up the stars), but if I had an achro I'd seriously consider this. In particular, I'd shoot the red and green separate from the blue. I just wonder if its focuser has enough range to bring the blue channel into focus.

 

BQ

On a 5" or 6" achromat, there is more than enough focus travel.  I tried this with the Celestron CR150 6" F/8 achromat I owned some years ago.  I seem to remember 5 or 10 millimeters at most, and I think actually less than that.  At that time stacking required resizing the channels so that the images would align.  Current stacking software manages that for you.  Should be possible.

 

Tom


Edited by Thomas A Davis, 19 June 2021 - 09:21 PM.


#14 BQ Octantis

BQ Octantis

    Fly Me to the Moon

  • *****
  • Posts: 5,198
  • Joined: 29 Apr 2017
  • Loc: Red Centre, Oz

Posted 19 June 2021 - 09:28 PM

At that time stacking required resizing the channels so that the images would align.  Current stacking software manages that for you.  Should be possible.

Crikey! Resizing?! That gives an idea of just how far out of focus the blue is! All very good to know…

 

BQ



#15 Thomas A Davis

Thomas A Davis

    Surveyor 1

  • *****
  • Posts: 1,708
  • Joined: 25 Mar 2005

Posted 19 June 2021 - 10:15 PM

Crikey! Resizing?! That gives an idea of just how far out of focus the blue is! All very good to know…

 

BQ

Just enough out of focus to be noticed.  On large achromats it is a very noticeable issue (such as the large Clark refractors such as the Lick 36" and Yerkes 40").  I understand the difference is focus is considerable.  Like I said, the stacking software handles it, so not really an issue today.  Years back I needed to do pixel math to determine how much to resize the blue channel.  That is where the noticeable difference was.

 

Didn't intend to scare anyone.

 

Tom



#16 BQ Octantis

BQ Octantis

    Fly Me to the Moon

  • *****
  • Posts: 5,198
  • Joined: 29 Apr 2017
  • Loc: Red Centre, Oz

Posted 19 June 2021 - 11:23 PM

Well I was only a little startled, and only for about 3 seconds at that. I then suddenly remembered that I've combined 200mm data with 750mm data without issue, so my fright abated quickly. And then I remembered…I don't even own an apo! lol.gif



#17 vidrazor

vidrazor

    Apollo

  • *****
  • Posts: 1,342
  • Joined: 31 Oct 2017
  • Loc: North Bergen, NJ

Posted 20 June 2021 - 01:55 AM

At the simplest level (without dealing with channel seps and such), I just brought in your full res M3 file (for example) into Photoshop (and I assume here you should have similar controls in GIMP) and I simply used the lens correction operator in the Camera Raw filters to remove the blue fringe. I then copied that layer, and I applied Smart Sharpen, then I adjusted that layer's luminosity mask to add transparency to the shadows and highlights of the sharpened layer so the black level noise and highlight edge enhancement were filtered out, so only the midtone data is sharpened. I then added a levels operator and balanced your black levels that were weak in red, and opened the gamma and compressed the shadow and highlight a bit, and got the image you see below (with your original for reference).

Hopefully that all made sense. smile.gif

Attached Thumbnails

  • m3 orig.jpg
  • m3.jpg

Edited by vidrazor, 20 June 2021 - 02:09 AM.

  • condensermike and the Elf like this

#18 vidrazor

vidrazor

    Apollo

  • *****
  • Posts: 1,342
  • Joined: 31 Oct 2017
  • Loc: North Bergen, NJ

Posted 20 June 2021 - 10:15 AM

One other thing I tried, a bit more convoluted, was to map the green channel into the blue, and again remove the fringing (which was now green smile.gif) I then brought that into Topaz Sharpen AI and saved it out of there and brought it back into the un-sharpened file. The Topaz app can do funky stuff with stars, so I mapped the large un-Topaz stars back in. I left the crunchiness in the smaller stars alone. What Topaz does is sometime good, sometimes not. I then adjusted the black balance again, but this time I brought the green and blue channels down instead of bringing the red channel up, and I think it made for a better color balance. This time I just opened the gamma and left the black and white points alone. So here's a comparison of the processes and the new hack. Again, hope these processes make sense. smile.gif

Attached Thumbnails

  • m3 grn-SharpenAI-focus.jpg
  • trio.jpg

Edited by vidrazor, 20 June 2021 - 10:17 AM.

  • condensermike 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: DSLR, Imaging



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics