Author Topic: NGC 6823 HaSHO  (Read 1756 times)

Offline rdryfoos

  • PixInsight Old Hand
  • ****
  • Posts: 372
    • View Profile
NGC 6823 HaSHO
« on: 2017 July 06 16:06:00 »
TOA 130 at F7.7, SBIG STT-8300 with 3nm Astrodon filters, 25 hours of data in 30 min subs (Ha-16, OIII 18, SII 10).  OIII very faint.  SII collected during 75% Moon
Calibrated with 400 bias, 68 flats and 33 darks.


Offline RickS

  • PTeam Member
  • PixInsight Jedi
  • *****
  • Posts: 1298
    • View Profile
Re: NGC 6823 HaSHO
« Reply #1 on: 2017 July 07 01:47:27 »
Nice work!

Offline rdryfoos

  • PixInsight Old Hand
  • ****
  • Posts: 372
    • View Profile
Re: NGC 6823 HaSHO
« Reply #2 on: 2017 July 08 18:32:16 »
Here's a crop in a more conventional SHO palette.  The original image is a bit over processed.  This was taken with a 5" scope.

Offline drmikevt

  • PixInsight Addict
  • ***
  • Posts: 112
    • View Profile
Re: NGC 6823 HaSHO
« Reply #3 on: 2017 July 12 08:11:01 »
Very nice - I like how 'natural' your image looks for a NB image. 

Offline rdryfoos

  • PixInsight Old Hand
  • ****
  • Posts: 372
    • View Profile
Re: NGC 6823 HaSHO
« Reply #4 on: 2017 July 13 16:40:27 »
Here is a more traditional SHO palette, still with green reduced with SCNR, but no boost in red or blue.  I switched the orientation (got sick of looking at this image I guess).

Offline rdryfoos

  • PixInsight Old Hand
  • ****
  • Posts: 372
    • View Profile
Re: NGC 6823 HaSHO
« Reply #5 on: 2017 July 13 16:41:55 »
Very nice - I like how 'natural' your image looks for a NB image.
  Thanks--I appreciate it!
Rodd

Offline rdryfoos

  • PixInsight Old Hand
  • ****
  • Posts: 372
    • View Profile
Re: NGC 6823 HaSHO
« Reply #6 on: 2017 July 14 07:07:00 »
I don't know what I was thinking with the last posting.  Wow....like a neon sign.  My apologies.  Here is a more toned down version.  This is tough data to process for me for some reason.  Can't seem to quite get it.

Offline drmikevt

  • PixInsight Addict
  • ***
  • Posts: 112
    • View Profile
Re: NGC 6823 HaSHO
« Reply #7 on: 2017 July 14 15:55:08 »
What did you do to go from the 'neon' one to the last one?  Don't be afraid to be detailed.

Offline rdryfoos

  • PixInsight Old Hand
  • ****
  • Posts: 372
    • View Profile
Re: NGC 6823 HaSHO
« Reply #8 on: 2017 July 14 19:25:21 »
What did you do to go from the 'neon' one to the last one?  Don't be afraid to be detailed.
  Well--It was just a matter of degree I think.  I used the same tools in the same order, but varied the settings.  The small incremental changes can really change an image dramatically.  It can be as simple as getting a better preview for background neutralization. or not being quite as aggressive with stretches.  I try to minimize noise control, and its easy to over sharpen--Here are my orders of operation for narrowband work after calibration and Mure Denoise

1) Combination into SHO palette using Channel Combination (no reason to use Pixel Math since I do not modify the %)
2) Dynamic crop (make icon)
3) DBE (make icon)
4) Background Neutralization
5) Color Calibration (Many do not do this with narrowband--I do, yields a very balanced histogram to start)
6)Stretch using Histogram Transformation (This step is critical and variations of approach result in different versions of the image.
   I use SCNR Green set to Average  Neutral to remove the green.  Sometimes I follow it with .30 Minimum neutral if I want more green out
7) Dynamic Crop Ha Master using icon

8) DBE Ha master using icon
9) Stretch Ha master so histogram is the same as the SHO image--I do it by eye and get it pretty close.  Not sure it needs to me precisely identical.
10) Protect SHO image while leaving stars exposed using a starmask created with the Starmask Tool and replace stars with Ha stars using Pixel Math
11) Extract Luminence and stretch using curves then use as inverted mask to protect nebulosity and replace back ground with Ha background image using pixel Math.
12) Extract luminence than using MSMT remove the R later (use 4 layer setting).  This results in a star mask that includes the tinyb stars
13) Stretch the MSMT star mask using curves to bring out as many tiny stars as possible--care must be taken not to bring out small nebular features
14) use MSLT star mask and replace tiny stars with Ha stars using Pixel Math (this is a tricky step that can be skipped--but I like white NB stars--even the tiny ones).
15)Using Channel Combination, insert Ha master as luminance into SHO image.  Sometimes I use a range mask and do a little Local Hist. equalization on Ha first, but not always

16)-TGV Denoise using lowest possible settings that eliminates noise without smearing image.  Trial and error--very important to get it right.  Its magic--you know it when you get it.

17) Selective sharpening using MSLT, Unsharp Mask and finished with Deconvolution with settings of .50 and .50 for standard deviation and shape to Luminance layer
18) Final TGV denoise to achieve smoothness without smearing

Then invariably follows numerous tweaks and undoings and more tweaks using Unsharp Mask, TGV denoise, Curves, Hist Trans and various range masks.  Sometimes, but not always I boost saturation just a tad--I mean a very little.  obviously I botched this step big time in the "neony" image! Its here that I ruin many images though I just never seem satisfied.

Here's an improved version.  I used slight tweaks of the curves,  histogram and a touch of noise control on the last posted image.



 

Offline rdryfoos

  • PixInsight Old Hand
  • ****
  • Posts: 372
    • View Profile
Re: NGC 6823 HaSHO
« Reply #9 on: 2017 July 15 05:55:45 »
Here is with a bit of deconvolution on the stars.  I like the effect in some areas, but not others.  Sort of on the fence
Rodd