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

Getting the right color for M27 using PixInsight

  • Please log in to reply
7 replies to this topic

#1 t-ara-fan

t-ara-fan

    Viking 1

  • -----
  • topic starter
  • Posts: 733
  • Joined: 20 Sep 2017
  • Loc: 50° 13' N

Posted 17 September 2019 - 12:37 AM

I have been working on my M27.

  • 19x 300" subs
  • ASI071MC Pro at -10°C
  • EdgeHD-8 with 0.7x reducer
  • ZWO OAG with ASI174MM mini guider.

 

Processing is done in PI.

 

Using STF, slightly dialed back, and various contrast enhancements and noise reductions I get this.  Stars are pretty white, but I have some nice detail in the middle IMHO.

 

eCLvWkS.jpg

 

Using ArcsinhStretch I get the pic below. The star in the middle is a blueish white dwarf, and the color reflects that.  But the red stuff which I assume is Ha is quite orange.

 

oFjuTe9.jpg

 

So ... what is the correct color for this nebula?  I like the nebula color more in the first pic, and the star colors more in the second pic.

 

Back when I used a DSLR I used rnc-color-stretch and always loved the colors.  But rnc-color-stretch does not like the super low ADUs that the sky is when using a cooled camera.

 

I put the cropped stack in a DropBox in case anyone has the time and skills to take a crack at this to pull out the best color possible.


Edited by t-ara-fan, 17 September 2019 - 12:44 AM.


#2 Ultranumbed

Ultranumbed

    Lift Off

  • -----
  • Posts: 1
  • Joined: 05 Jun 2018

Posted 17 September 2019 - 01:28 AM

JkuO7Xn.jpg

 

 

This is right after DBE -> BN - > CC -> HistogramTransformation (Applied  STF Autostretch).

 

 

UQ8YMMD.jpg

 

I did some quick processing (no noise reduction), and ended up with this. The color shouldn't change much if you perform DBE -> BN -> CC.


Edited by Ultranumbed, 17 September 2019 - 01:31 AM.

  • t-ara-fan likes this

#3 ChristopherBeere

ChristopherBeere

    Vostok 1

  • -----
  • Posts: 103
  • Joined: 24 Jul 2019
  • Loc: London UK

Posted 17 September 2019 - 04:48 AM

I had a quick go with the data to see if i could achieve the goal of presenting your preferred nebula color from the standard stretch and the star color from the ArcsinH stretch.

 

 

FdJj2vf.jpg

 

 

 

The way i did it was to process the stars and nebula seperately and then add them back together. 


Edited by ChristopherBeere, 17 September 2019 - 05:23 AM.

  • t-ara-fan and Giorgio Ferrari like this

#4 FiremanDan

FiremanDan

    Aurora

  • *****
  • Posts: 4771
  • Joined: 11 Apr 2014
  • Loc: Virginia

Posted 17 September 2019 - 09:02 AM

Have you tried the photometric calibration?

 

I tried but I didn't have the focal length and pixel scale and it was cropped so it wouldn't work. I used that with my M13 to get great color and accurate color. I felt like M13 was too blue in most people's images. Globular clusters are older stars and blue stars are young super hot stars.

Anyways, I ran it through ABE and basic color calibration. Arcsinh stretch. I also tried a trick in PS where you open up curves and use the grey eyedropper tool to select a dark background and it'll usually snap the RGB curves to their ideal spots. I couldn't get anything better than I got in PI with the basic color calibraiton. I used one of the brighter clipped stars as my white reference and a small section of dark background for the background reference and it looks like a good M27 color to me. Then I went into PS and upped the vibrance a very small amount and upped the contrast just a touch.
For next to no processing, I think that's a nice looking M27!
 

t-ara-fan M27

Then I decided to run it through HDRmultiscale (6 to lightness) and LHE with a mask via RangeSelection to get some of the details in the core to pop out.

 

t-ara-fan M27 LHE

  • t-ara-fan likes this

#5 NorthField

NorthField

    Viking 1

  • *****
  • Posts: 932
  • Joined: 01 Jun 2017
  • Loc: SW Missouri

Posted 17 September 2019 - 10:54 AM

i tried pcc a few different times but it always turned the entire image blue.. i've never seen that before from pcc, weird.... anyway, just normal bn and cc worked fine... mix of arcsinh and stf and a little lhe
 
if you can get more time on this sucker, you'll have a real WINNER!!
 
tarafans M27

 


  • t-ara-fan likes this

#6 t-ara-fan

t-ara-fan

    Viking 1

  • -----
  • topic starter
  • Posts: 733
  • Joined: 20 Sep 2017
  • Loc: 50° 13' N

Posted 17 September 2019 - 12:41 PM

I had a quick go with the data to see if i could achieve the goal of presenting your preferred nebula color from the standard stretch and the star color from the ArcsinH stretch.

 

 

The way i did it was to process the stars and nebula seperately and then add them back together. 

Looks good, I guess I will have to learn more about masks. How did you do it?  I am definitely a PI noob, slowly learning.

 

Did you make a LUM mask, then remove the stars to leave the nebula?  Remove them with a brush in photoshop? Then invert the mask and STF the nebula?

Take the same mask, remove the nebula, then ArcsinhStretch the stars?

How do you add the two together? PixelMath?



#7 t-ara-fan

t-ara-fan

    Viking 1

  • -----
  • topic starter
  • Posts: 733
  • Joined: 20 Sep 2017
  • Loc: 50° 13' N

Posted 17 September 2019 - 12:45 PM

Have you tried the photometric calibration?

 

I tried but I didn't have the focal length and pixel scale and it was cropped so it wouldn't work t.

 

I really like the colors you brought out.

I think I tried PMCC, but the PlateSolve failed. Not sure why. It has always worked in the past.

EdgeHD-8 at 0.7x = 1422mm FL.  ASI071MC Pro 4.78um pixels.

 

And thanks to all for your efforts.


Edited by t-ara-fan, 17 September 2019 - 12:46 PM.

  • FiremanDan likes this

#8 ChristopherBeere

ChristopherBeere

    Vostok 1

  • -----
  • Posts: 103
  • Joined: 24 Jul 2019
  • Loc: London UK

Posted 17 September 2019 - 01:01 PM

Looks good, I guess I will have to learn more about masks. How did you do it?  I am definitely a PI noob, slowly learning.

 

Did you make a LUM mask, then remove the stars to leave the nebula?  Remove them with a brush in photoshop? Then invert the mask and STF the nebula?

Take the same mask, remove the nebula, then ArcsinhStretch the stars?

How do you add the two together? PixelMath?

 

Just did a basic manual stretch and balanced the colors with HTF to get the Nebula looking about right.

 

Then I used the Starnet++ module for PI to extract the nebula.

 

https://sourceforge....PI.zip/download

 

Save this result as M27 Nebula.

 

Then i opened the original file and stretched it with ArcsinH to get the stars how i wanted them. Save this as M27 Stars.

 

Next, use Starnet++ to remove the nebula from this image. Save this result as M27 Nebula Remove.

 

Open M27 Stars in photoshop and create another layer. Open M27 Nebula Remove and dump it in as the second layer. Use difference blend mode to subtract the nebula from the stars and then flatten it.

 

Now you have the ArcsinH stretched star field isolated. Add a new layer and load in M27 Nebula. Now use screen blend mode to add the stars and nebula back together.

 

Job done.

 

Im a Pixinsight noob as well so i have to rely on Photoshop to do a bunch of stuff, im sure one of the guys will be able to tell you how to do all of this in PI though.


Edited by ChristopherBeere, 17 September 2019 - 01:13 PM.

  • t-ara-fan 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






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics