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

How to achieve great star color?

  • Please log in to reply
9 replies to this topic

#1 MakNewtMan

MakNewtMan

    Sputnik

  • -----
  • topic starter
  • Posts: 48
  • Joined: 28 Nov 2018

Posted 26 May 2019 - 09:31 AM

I’ve noticed in my image processing the stars in my image end up white (clipped with no color). Would doing a masked stretch fix this or does exposure length play a part as well? Should I take a separate set of subs that are much shorter than my exposures of my main object of interest and then stack those and blend with my other stacked image? Trying to find the best way to attain that beautiful blue and orange colored stars I’ve seen in so many astrophotos. I appreciate any tips or advice.

#2 fmeschia

fmeschia

    Apollo

  • *****
  • Posts: 1017
  • Joined: 20 May 2016
  • Loc: Mountain View, CA

Posted 26 May 2019 - 11:29 AM

How do you process your images? Are you using PixInsight?

Francesco


Edited by fmeschia, 26 May 2019 - 11:30 AM.


#3 MakNewtMan

MakNewtMan

    Sputnik

  • -----
  • topic starter
  • Posts: 48
  • Joined: 28 Nov 2018

Posted 26 May 2019 - 12:28 PM

Hi, I use photoshop CC to process my images

#4 fmeschia

fmeschia

    Apollo

  • *****
  • Posts: 1017
  • Joined: 20 May 2016
  • Loc: Mountain View, CA

Posted 26 May 2019 - 01:16 PM

Ok. I don’t use Photoshop so I don’t know the tools there, but the general idea is that you lose color when stretching saturates all channels which gives you pure white. In PixInsight there are tools that apply various levels of masking to the star cores and halos to preserve the relative intensity of the channels while stretching, i.e. they preserve the star colors. I guess you could try achieving a similar result in Photoshop with a mask layer?

Francesco


  • elmiko likes this

#5 17.5Dob

17.5Dob

    Fly Me to the Moon

  • ****-
  • Posts: 5083
  • Joined: 21 Mar 2013
  • Loc: Colorado,USA

Posted 26 May 2019 - 08:29 PM

Mark (sharkmelly) wrote a PS action for ArcSinh stretch a while back. You might try that.

Even using all of the tools and plugins I have in PS, moving to PI and using ArcSinh stretch was an eye-opener.

Best PS attempt

38105077134_76478003bd_c.jpg

2 clicks in PI ( PCC and ArcSinh stretch)

33541740568_b8cb1ec05e_c.jpg


Edited by 17.5Dob, 26 May 2019 - 08:30 PM.

  • mikewayne3, bobzeq25 and fmeschia like this

#6 PRejto

PRejto

    Mariner 2

  • -----
  • Posts: 248
  • Joined: 11 Feb 2007
  • Loc: Sydney, Australia

Posted 27 May 2019 - 10:59 AM

In answer to your first question, is it sub length or processing, the answer is in the original data. Anything that is saturated cannot show color. I don't know what image capture software you use but you should be able to mouse over stars and read the adu values. With very bright stars you might just saturate no matter what you do but if the bulk of background stars are less than 65,000 adu you ought to see color. The initial stretch is extremely important. I use both CCDStack and/or the Pixinsight Histogram Transformation tool when going non-linear. It is extremely important to keep things gray and dim looking initially. Default stretches are often far too stretched! As you stretch you can see the stars brighten (and expand). Really watch it and just stop or reverse stretching. You might need a star mask, maybe not. It is often surprising to me that even though the initial stretch looks to be far from enough the final image is quite good.

 

Peter



#7 MakNewtMan

MakNewtMan

    Sputnik

  • -----
  • topic starter
  • Posts: 48
  • Joined: 28 Nov 2018

Posted 28 May 2019 - 08:20 AM

That’s helpful Peter, thank you. Your image comparisons between PS and Pixinsight are night and day, wow. I’ve been on the fence about moving to pixinsight but this illustration definitely makes me want to make that change sooner. How’s the learning curve on pixinsight?

#8 the Elf

the Elf

    Surveyor 1

  • *****
  • Posts: 1599
  • Joined: 06 Sep 2017
  • Loc: Germany

Posted 28 May 2019 - 09:11 AM

Yep, arcsinh stretch is the way to go. If you want to use PS, convert the image to L-a-b. Apply a curves operation to a and b, using a straight line through the center, same for a and b. The technique is described in Charles Bracken's Deep-Sky Imaging Primer on page 145 in the first edition, the section is called "Boosting Saturation in LAB". Back in RGB you can do an additional saturation.

Grab some popcorn and watch this: https://www.youtube....h?v=mvuMrHEWaS8


Edited by the Elf, 28 May 2019 - 09:12 AM.


#9 Pete_xl

Pete_xl

    Viking 1

  • -----
  • Posts: 677
  • Joined: 29 Jan 2015
  • Loc: Germany

Posted 28 May 2019 - 09:18 AM

Mark (sharkmelly) wrote a PS action for ArcSinh stretch a while back. You might try that.

Even using all of the tools and plugins I have in PS, moving to PI and using ArcSinh stretch was an eye-opener.

Best PS attempt

38105077134_76478003bd_c.jpg

2 clicks in PI ( PCC and ArcSinh stretch)

33541740568_b8cb1ec05e_c.jpg

 

 

That’s helpful Peter, thank you. Your image comparisons between PS and Pixinsight are night and day, wow. I’ve been on the fence about moving to pixinsight but this illustration definitely makes me want to make that change sooner. How’s the learning curve on pixinsight?

I have hesitated a while but seeing your final conclusion I feel I must give a comment. The comparison above is not representative nor in any way objective. By the way Mark has also disclosed Photoshop actions for his arcsinh stretch method. There are numerous examples of great images that are made in Photoshop. I use Photoshop myself in combination with Astro Pixel Processor. I have also a Pixinsight license. At least for me I can say that think I will never be able to get results out of Pixinsight that can compete with what APP + Photoshop CC  me. And PI is the opposite of ergonomic.

 

Photoshop and Pixinsights base on a different concept and in both cases the quality of the result depends on the skills of the person who pushes the buttons.


  • 44maurer likes this

#10 the Elf

the Elf

    Surveyor 1

  • *****
  • Posts: 1599
  • Joined: 06 Sep 2017
  • Loc: Germany

Posted 28 May 2019 - 11:39 AM

 

the result depends on the skills of the person who pushes the buttons

This is true for almost anything in life. Anyone doing whatever needs to find a tool that fits his or her way of working and thinking.Of course there are bad tools and bad programms. But as soon as you reach a certain qualitiy level it is far more important to find something that fits your needs. Apart from being "your" tool there are different oppinions about what looks good. Also if you know a software you come to better results compared to starting with a new tool you have never used before. I guess that most people agree on this: PI is designed for astro photography and so is APP but PS is build for a far wider scope of landscape and portrait, etc.

It may be me, but for my hobby I listen to my feelings. If a program makes me worry I get another one because I want to enjoy my hobby. Even if all experts say it is the better tool I don't want to be frustrated. Try several programs, get the one the feels best, take some steps and if it turns out to suck constantly, get another one.


  • 44maurer and B 26354 like 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