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

Pulling my hair out -- and there's not much left!

  • Please log in to reply
13 replies to this topic

#1 Calypte

Calypte

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1,556
  • Joined: 20 Mar 2007
  • Loc: Anza, California

Posted 26 February 2020 - 04:35 AM

This was shot with a standard Canon 6D through a TEC140ED refractor.  This is typical of what I get.  I thought I'd be getting real pictures.  They probably wouldn't be perfect, but the images would be recognizably close to what they are supposed to be.  Color balance in the camera is set for daylight.  Real daytime snapshots from this camera look fine. 

 

I don't know what to do to prevent this.  I've tried turning off the webcam in my observatory, which sprays out IR at night.  I've tried covering the LED on the guide camera.  I replaced the camera's AC adapter with the standard Canon battery, just thinking that maybe the adapter caused some odd interference in the camera.  Nothing I've tried has made the slightest difference.

Attached Thumbnails

  • M78_Canon-6d_20200224_small.jpg

Edited by Calypte, 26 February 2020 - 04:37 AM.


#2 sg6

sg6

    Fly Me to the Moon

  • *****
  • Posts: 7,257
  • Joined: 14 Feb 2010
  • Loc: Norfolk, UK.

Posted 26 February 2020 - 04:56 AM

Is the camera modified ?

Although I see you say "standard"

 

No mention of "darks".

Likely going to need a few. Basically take camera, put covers on it, set whatever timer for the same length exposure and say 1/3 to 1/2 as many exposures, put in fridge, press Go button, close door, have coffee.

 

The images of what amounts to nothing are added to DSS and that could set the background level to something dark.

 

DSLR's have a set "grey" level where the camera will set a default level of light and yours may be doing that, since the background is dark it could be raising everything up to this "grey" level. So the background is brigthened, effectively from black to grey,  and everything else is also.

 

Darks are easy to take all you need is a fridge, well that is what I used. No effort, no focus, no anything.

 

Is there a "Night" setting on the camera, BUT this could simply brighten everything. Depends on the specification of what it was decided it does. It may not make night skies dark/black. I suspect a few darks are what you need.

 

Just thought: Are you taking a series of exposures and stacking, or are you trying a single photograph? Post doesn't seem to make that clear.


Edited by sg6, 26 February 2020 - 04:59 AM.

  • Calypte and Xeroid like this

#3 pyrasanth

pyrasanth

    Mercury-Atlas

  • *****
  • Posts: 2,513
  • Joined: 08 Jan 2016

Posted 26 February 2020 - 05:22 AM

I think the OP has somewhat enhanced expectations with regard to astro photography. I have never seen an image come out of any camera on an astronomy target without a colour cast. These are very easy to correct in post processing.

 

Photoshop has plugins like the excellent gradient exterminator or astro flat pro or even some of the more basic image processing packages for astro photography all have the ability to remove background gradients. These are nearly always created where one colour channel is slightly out of balance and are the norm and nothing extraordinary.

 

The image you have presented will clean up very nicely in any of these post processing processes- it really is about understanding what is going on & recognizing that astro photography is very different from terrestrial pictures.



#4 james7ca

james7ca

    Cosmos

  • *****
  • Posts: 8,022
  • Joined: 21 May 2011
  • Loc: San Diego, CA

Posted 26 February 2020 - 06:39 AM

Kind of looks like regular light pollution. However, if this was shot from Anza and if the lights from Anza aren't really that bad (it's a fairly small city) then shooting to the south (M78) should be fairly dark. Well, that assumes that this was done with M78 near the meridian because if you shot this when Orion was low in the western sky then all bets are off.

 

That said, I've never gotten a good image of M78 even though some sources say it is one of the brightest reflection nebulae in the sky (reflection being different than an emission nebulae). However, my skies have red/orange zone light pollution and my southern horizon is the worst part of my sky.

 

Also, I assume that this was shot in RAW mode and not JPEG or TIFF. The latter two shouldn't cause any particular problems with color, but you want to capture RAW files for all of your deep space photography.

 

In any case, yes, you are always going to have some color shifts and color casts when doing DSO imaging and you'll need to "fix" those in post processing.

 

Below is a sample I got after I did a simple DynamicBackgroundExtraction using PixInsight (on the OP's uploaded JPEG, not a good place to start). Then I did a mild histogram stretch and LocalHistogramEqualization (the latter to brighten the nebula, not to change the color).

Attached Thumbnails

  • M78_Orange_Color_Cast_DBE.jpg

Edited by james7ca, 26 February 2020 - 06:40 AM.

  • Calypte, sharkmelley and Kevin_A like this

#5 Blackbelt76

Blackbelt76

    Mariner 2

  • -----
  • Posts: 230
  • Joined: 14 Oct 2019

Posted 26 February 2020 - 07:12 AM

Color balance in the camera is set for daylight.

 

Color balance/ white balance is meaningless if you shoot RAW, which is what you should be doing for any AP images.

RAW capture does not lock in white balance and is easily adjusted in post processing.

 

AP post processing has quite a learning curve as I am learning compared to terrestrial photography.

Start by adjusting your WB on ONE light frame; then batch process the remaining frames BEFORE stacking.


  • Juno18 likes this

#6 sharkmelley

sharkmelley

    Mercury-Atlas

  • *****
  • Posts: 2,919
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 26 February 2020 - 07:32 AM

What you are seeing is the typical brown/orange light pollution.  M78 is a very faint and challenging object to image, so it will almost always be dominated by the light pollution.

 

Light pollution needs to be subtracted from the data, preferably while it is still linear. 

 

Mark


Edited by sharkmelley, 26 February 2020 - 07:38 AM.

  • Ishtim likes this

#7 erictheastrojunkie

erictheastrojunkie

    Viking 1

  • *****
  • Posts: 618
  • Joined: 15 Jul 2016
  • Loc: Salt Lake City

Posted 26 February 2020 - 09:21 AM

As others have said, looks completely normal to me, just do a simple background extraction and then stretch the image.  


  • csridgway likes this

#8 17.5Dob

17.5Dob

    Fly Me to the Moon

  • ****-
  • Posts: 5,884
  • Joined: 21 Mar 2013
  • Loc: Colorado,USA

Posted 26 February 2020 - 11:16 AM

This was shot with a standard Canon 6D through a TEC140ED refractor.  This is typical of what I get.  I thought I'd be getting real pictures.  They probably wouldn't be perfect, but the images would be recognizably close to what they are supposed to be.  Color balance in the camera is set for daylight.  Real daytime snapshots from this camera look fine. 

 

I don't know what to do to prevent this.  I've tried turning off the webcam in my observatory, which sprays out IR at night.  I've tried covering the LED on the guide camera.  I replaced the camera's AC adapter with the standard Canon battery, just thinking that maybe the adapter caused some odd interference in the camera.  Nothing I've tried has made the slightest difference.

That is completely normal. Even from a Bortle 1 zone, airglow is going to cause a color cast in your image. You need to get a reasonable stack, and then use a background neutralizer.


  • Calypte likes this

#9 Calypte

Calypte

    Surveyor 1

  • -----
  • topic starter
  • Posts: 1,556
  • Joined: 20 Mar 2007
  • Loc: Anza, California

Posted 26 February 2020 - 01:47 PM

Thank you for your replies.  This is a standard, stock Canon 6D.  In fact I have a Hutech-modified camera, and I was getting results sort of like this, so I gave up on it -- for the time being.  But this one is absolutely stock.  The original of this shot is RAW.  There's been a learning curve of coordinating CCD Commander and MaxIm DL 6.21 to do this, because, although MaxIm can do RAW, it doesn't really want to.  MaxIm's preferred format is to put the RAW data into a FITs.  Anyway, the original of this is pure RAW.  The pic I posted is a JPEG derived from a TIFF, but this gets the point across.  I have 13 frames available.

 

My sky is (I've been told) Bortle 4, although I'd give it more of a 3.5.  The best nights are when clouds cover the lowlands.

 

I know this isn't what most of you do, but I've been trying to use Tony Hallas's methods as he describes here:

https://www.youtube.com/watch?v=PZoCJBLAYEs&t=808s

I've swapped some messages with Tony about this.  I think I'm getting different results because he is (was) mostly doing large Milky Way-scapes, not imaging through a telescope.

 

Mr. james7ca: I have PixInsight, but DBE looks complicated, so I haven't really investigated it.  I know that Adam Block, even before he converted completely to PixInsight, was using DBE.


Edited by Calypte, 26 February 2020 - 01:53 PM.


#10 fmeschia

fmeschia

    Apollo

  • *****
  • Posts: 1,236
  • Joined: 20 May 2016
  • Loc: Mountain View, CA

Posted 26 February 2020 - 02:36 PM

Try ABE if DBE is too complicated. It will lift most of that that orange veil, allowing you to see what’s behind. 


Edited by fmeschia, 26 February 2020 - 02:36 PM.

  • Michael Covington and Swordfishy like this

#11 bobzeq25

bobzeq25

    Hubble

  • *****
  • Posts: 18,986
  • Joined: 27 Oct 2014

Posted 26 February 2020 - 05:45 PM

Thank you for your replies.  This is a standard, stock Canon 6D.  In fact I have a Hutech-modified camera, and I was getting results sort of like this, so I gave up on it -- for the time being.  But this one is absolutely stock.  The original of this shot is RAW.  There's been a learning curve of coordinating CCD Commander and MaxIm DL 6.21 to do this, because, although MaxIm can do RAW, it doesn't really want to.  MaxIm's preferred format is to put the RAW data into a FITs.  Anyway, the original of this is pure RAW.  The pic I posted is a JPEG derived from a TIFF, but this gets the point across.  I have 13 frames available.

 

My sky is (I've been told) Bortle 4, although I'd give it more of a 3.5.  The best nights are when clouds cover the lowlands.

 

I know this isn't what most of you do, but I've been trying to use Tony Hallas's methods as he describes here:

https://www.youtube.com/watch?v=PZoCJBLAYEs&t=808s

I've swapped some messages with Tony about this.  I think I'm getting different results because he is (was) mostly doing large Milky Way-scapes, not imaging through a telescope.

 

Mr. james7ca: I have PixInsight, but DBE looks complicated, so I haven't really investigated it.  I know that Adam Block, even before he converted completely to PixInsight, was using DBE.

Below are illustrative pictures.  A single frame (stretched so you can see it).  (the color of the color cast is completely irrelevant)  The unstretched stack, as (correctly) seen when loaded into a processing program.  The final image.  I trust that gives you hope.  <smile>

 

PI is not for everyone.  You may see that it's the "best", but that's only true when you spend serious time learning it.  I have hundreds of hours in it.  DBE is actually one of the simpler operations.

 

To make PI the best you have to use the complexity, use the many parameter adjustments.  The complexity, as far as results are concerned, is not a bug, it's a feature.

 

Astro Pixel Processor also both stacks and processes.  Has an excellent gradient reduction tool.  Is vastly easier to learn and use.  Gives fine results.  Consider it hair preservation.  <smile>

 

https://www.astropixelprocessor.com/

 

The Hallas method decimates the data.  He tried hard to promote it, by and large the imaging community passed, they work hard to get good data.  Good critique here, from a major, recognized expert in astrophotography.

 

https://www.cloudyni...r/#entry6306310

 

Rosette sub.jpg

 

Rosette(32l,f,b,d).jpg

 

RosetteV3.1-small.jpg


Edited by bobzeq25, 26 February 2020 - 05:51 PM.

  • Swordfishy likes this

#12 sharkmelley

sharkmelley

    Mercury-Atlas

  • *****
  • Posts: 2,919
  • Joined: 19 Feb 2013
  • Loc: UK

Posted 26 February 2020 - 07:35 PM

 

I know this isn't what most of you do, but I've been trying to use Tony Hallas's methods as he describes here:

https://www.youtube.com/watch?v=PZoCJBLAYEs&t=808s

I've swapped some messages with Tony about this.  I think I'm getting different results because he is (was) mostly doing large Milky Way-scapes, not imaging through a telescope.

 

If you are determined to use the Tony Hallas / Roger Clark approach then why not do as they suggest to remove the light pollution i.e. levels and curves.  Roger Clark has detailed instructions on his ClarkVision site - better than the Tony Hallas video - he even goes on to describe a method of removing the remaining gradients by making feathered selections with levels.

 

I don't support the Hallas/Clark approach because it's not based on sound mathematics.  But a lot of people do use these methods to get quick results.

 

Mark


Edited by sharkmelley, 27 February 2020 - 03:22 AM.


#13 Juno18

Juno18

    Vostok 1

  • *****
  • Posts: 100
  • Joined: 07 Dec 2018
  • Loc: Long Beach, Mississippi, USA

Posted 26 February 2020 - 07:56 PM

This is a two second process of the 45kb image that I downloaded.

 

Opened up in Photoshop Elements, Enhance drop down menu, Adjust Lighting, Levels, Click the Set Gray Point dropper, click the background. Looks roughly 'Normal".

 

Just rough, but the red doesn't seem to be a big issue.

 

 

Attached Thumbnails

  • post-24654-0-30090700-1582709821 new.jpg


#14 james7ca

james7ca

    Cosmos

  • *****
  • Posts: 8,022
  • Joined: 21 May 2011
  • Loc: San Diego, CA

Posted 26 February 2020 - 09:42 PM

DynamicBackgroundExtraction (DBE) has a lot of options and settings but for your image (above) I used my standard approach, click on the image (to select), click on the button to create samples, select subtract for the correction method, and then execute the process. That's four mouse clicks (that actually do something, a few more to reveal the GUI) and I left everything else at the defaults (I sometimes tweak a few of the settings, but not for this example).

 

I'm pretty sure there are also instructional videos on using this tool, try Harry's Astro Shed:

 

  https://www.harrysas...nsighthome.html

 

Or, specifically (on DBE):

 

  https://www.harrysas...ids/dbe/dbe.mp4

 

That said, there are other ways to get similar results (quickly).


Edited by james7ca, 26 February 2020 - 09:44 PM.



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