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

Please help me with a colour assessment of M51

  • Please log in to reply
13 replies to this topic

#1 pyrasanth

pyrasanth

    Vanguard

  • *****
  • topic starter
  • Posts: 2,497
  • Joined: 08 Jan 2016

Posted 18 February 2020 - 06:16 PM

Whilst I'm waiting for the new camera I took a few subs of M51 with the RASA 11 & QHY168C. Is the image too blue?

 

I'm using this image to practice how to get the best colour I can for future processing especially for the M81/M82 Big data image. I think I've hit on a technique that works quite well but I'd like the forums opinion as to how they think the colour is on this image. I'm trying to steer away from PCC or any of the more advanced correction techniques for now.

 

The image below has had no processing other than the new colour technique which I'm working with but is quite simply using multiple archsinh stretches and not the masked stretch which can leave an image really dark. The image has no L channel or noise reduction added or noise removed.

 

The image is 680 1 minute subs under Bortle 8/9 skies. I intend to capture a full block of L data when I get the new camera and blend into this image as the RGB data is quite nice. I will remove the stars and process separately when I get the L data.

 

As always, let me know your thoughts.

 

Happy imaging to you all.

 

Best wishes

Mark

 

M51_doubleARCSIN_web.jpg


Edited by pyrasanth, 18 February 2020 - 06:17 PM.

  • happylimpet and Astroman007 like this

#2 Jim Waters

Jim Waters

    Soyuz

  • *****
  • Posts: 3,806
  • Joined: 21 Oct 2007
  • Loc: Phoenix, AZ USA

Posted 18 February 2020 - 06:51 PM

Real nice image. Good detail and guiding. The colors are good but you should back off on the contrast. The colors intensity is a bit too high.

 

What PI steps did you follow?


Edited by Jim Waters, 18 February 2020 - 06:55 PM.

  • pyrasanth likes this

#3 Astroman007

Astroman007

    Voyager 1

  • *****
  • Posts: 10,893
  • Joined: 27 Oct 2017
  • Loc: Northern Ontario, Canada

Posted 18 February 2020 - 07:05 PM

I like it just the way it is.

 

But to those whose preferences do not favor the blue end of the spectrum, I could see what you're getting at.



#4 sunnyday

sunnyday

    Aurora

  • *****
  • Posts: 4,951
  • Joined: 23 Sep 2019
  • Loc: Ottawa,Canada

Posted 18 February 2020 - 08:32 PM

really a very nice catch
the detail, the color, I think everything is well balanced, even the continuous monitoring is good.



#5 OhmEye

OhmEye

    Messenger

  • -----
  • Posts: 454
  • Joined: 15 Sep 2019
  • Loc: Western NY Southern Tier

Posted 18 February 2020 - 11:07 PM

I like the contrast and color balance also, but my taste does skew more blue than may be popular.



#6 happylimpet

happylimpet

    Skylab

  • *****
  • Posts: 4,450
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 19 February 2020 - 03:14 AM

Looks lovely. I wonder if its sliiiiightly too blue, but nothing to complain about!


  • pyrasanth likes this

#7 pyrasanth

pyrasanth

    Vanguard

  • *****
  • topic starter
  • Posts: 2,497
  • Joined: 08 Jan 2016

Posted 19 February 2020 - 04:10 AM

Real nice image. Good detail and guiding. The colors are good but you should back off on the contrast. The colors intensity is a bit too high.

 

What PI steps did you follow

Thanks Jim,

 

I've not looked at the contrast- it came out pretty much as is. The processing steps for the 680 subs are indicated below:

 

1. Weighted batch pre-processing with automatic selection of best image for registration & integration. VNG debayer & cosmetic correction not applied (still working on finding the best settings for this)

2. Background neutralization & colour calibration

3. Dynamic background correction

4. SCNR green removal

5. Gentle ABE

6. ArcsinH stretch at an initial setting of 50

7. Archsinh stretch repeated to suit your preference

8. Rough adjustment of colours and levels using histogram or curves transformation

9. Export to photoshop for a few levels tweaking.

 

That is all the basis steps I used for this image. 


Edited by pyrasanth, 19 February 2020 - 04:16 AM.


#8 pyrasanth

pyrasanth

    Vanguard

  • *****
  • topic starter
  • Posts: 2,497
  • Joined: 08 Jan 2016

Posted 19 February 2020 - 04:13 AM

Looks lovely. I wonder if its sliiiiightly too blue, but nothing to complain about!

Thanks happylimpet- We are having a bit of a blue consensus. I feel that the image is a tad too blue but others like the blue. I might try a bit of a halfway house. The image will change when I can capture the L. I'm saving for the new camera now- an IX455 sensor- but its quite an investment so its stretching the available budget but worth it in the long run.



#9 pyrasanth

pyrasanth

    Vanguard

  • *****
  • topic starter
  • Posts: 2,497
  • Joined: 08 Jan 2016

Posted 19 February 2020 - 04:27 AM

Thanks for all your kind and thoughtful observations. I'm going to back off the blue slightly and perhaps an adjustment of contrast might have the desired effect. However the image currently does not have an implicit L channel applied so I believe there is a lot more detail to bring out as colour alone does not carry the degree of spatial information.

 

The subs are unguided at 60 second exposures and appears to be just right for my sky conditions and this target. When I blink the subs I can see enough movement to suggest that a drizzle might work when I get the IMX455 camera as I suspect I will be under-sampled with a 3.7 nm pixel size at 620 mm focal length.

 

I did obtain the drizzle information for the colour subs at the time of registration. I may have a look to see if this provides any benefit to the image detail but expect the most effect on the L channel to be captured.

 

I'm going to look very carefully at the 680 subs I've captured and select a sample of the best with the roundest stars to produce a separate integration for the stars. I can then remove the over-saturated stars and replace with a better star-field.

 

I don't really know how many subs I will need for the L channel I'm mindful that even at 680 subs on the OSC the image is really bright so I might shoot for 500 on the new camera to make things really zing!

 

I will post updates as I work through this flow.

 

Happy imaging to you all.

 

Best wishes

Mark


Edited by pyrasanth, 19 February 2020 - 04:33 AM.

  • happylimpet likes this

#10 happylimpet

happylimpet

    Skylab

  • *****
  • Posts: 4,450
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 19 February 2020 - 04:58 AM

I must say your recent images with that RASA really seem to have come on a level! Nice one.


  • pyrasanth likes this

#11 pyrasanth

pyrasanth

    Vanguard

  • *****
  • topic starter
  • Posts: 2,497
  • Joined: 08 Jan 2016

Posted 19 February 2020 - 05:34 AM

I must say your recent images with that RASA really seem to have come on a level! Nice one.

Thank you- that's a nice compliment. I'm getting lots of practice. I had to upgrade the computers to handle the massive increase in data- 60 second subs really take a lot of processing. I envy the long focal length users who process 10 minute subs!



#12 happylimpet

happylimpet

    Skylab

  • *****
  • Posts: 4,450
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 19 February 2020 - 05:54 AM

Thank you- that's a nice compliment. I'm getting lots of practice. I had to upgrade the computers to handle the massive increase in data- 60 second subs really take a lot of processing. I envy the long focal length users who process 10 minute subs!

This is a concern of mine. I would like in due course to use PI, but I hear its very sluggish with many subs. DeppSkyStacker laughs at 1000 or even 10,000 subs, but do you think PI can do this on a modestly fast PC with 16GB 3200MHz RAM and AMD Ryzen 7 2700X 3.7GHz Octa Core (Socket AM4)?



#13 pyrasanth

pyrasanth

    Vanguard

  • *****
  • topic starter
  • Posts: 2,497
  • Joined: 08 Jan 2016

Posted 19 February 2020 - 06:39 AM

This is a concern of mine. I would like in due course to use PI, but I hear its very sluggish with many subs. DeppSkyStacker laughs at 1000 or even 10,000 subs, but do you think PI can do this on a modestly fast PC with 16GB 3200MHz RAM and AMD Ryzen 7 2700X 3.7GHz Octa Core (Socket AM4)?

I don't think it is the fault of PI- if "fault" is the right word. Most computers are setup for normal everyday tasks- I doubt if many are optimal right out of the box for the hammering they would get from astro processing. I think we need to think out of the box and look at building computers for PI if that is your choice of software moving forward.

 

The computer I now use can happily integrate a days production of subs- that is about 400 & even happily cope with twice that many but it is built to handle that type of processing and the demands around memory & hard drive space.

 

The situation should not get worse moving forward as hardware becomes increasingly powerful however earlier systems may suffer as they come under increasingly demanding workloads from larger sensors and the brutal speed of CMOS vs CCD. 

 

The system you have quoted will easily tackle a couple of hundred PI subs using batch processing provided the swap file space has been correctly allocated. PI is not sluggish on the right system- it can be very fast. The sweet spot for RAM is always going to be 64 GB- that is the current weakness of your system- upgrade your memory and you will be well placed to integrate 500 subs plus.


Edited by pyrasanth, 19 February 2020 - 06:49 AM.

  • happylimpet likes this

#14 leviathan

leviathan

    Viking 1

  • *****
  • Posts: 674
  • Joined: 29 Nov 2011
  • Loc: Azerbaijan

Posted 19 February 2020 - 09:25 AM

Yes, image is slightly blueish, especially the background (space). Fantastic details nevertheless !




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