Fireworks Galaxy Processing Request [Deep Sky] Processing techniques · Fabian Butkovich · ... · 10 · 411 · 9

FabianButkovich 0.90
...
· 
·  1 like
·  Share link
I haven't been too pleased with any variation of final edit I've created from my 2024 data on NGC 6946, The Fireworks Galaxy. I figured I would let others have a go at processing it to see if anyone else can showcase this beautiful target better. This is my own recent re-edit (RevC), with the following PixInsight workflow/process.


NGC 6946 (Fireworks Galaxy) 2024



image.png


The final image is actually a photoshop mean blend with my previous revision (RevB) since I liked the color of the newer version better but wanted to keep the luminosity of the previous version. This was my actual final image out of PixInsight


image.png


I did use drizzle2X in PixInsight WBPP and then down sampled later in post to try and increase the SNR.

Here is the link to all of my data, including flat calibration frames:

NGC 6946 2024

I have also included my master light out of WBPP incase any of you wish to use that instead, the master light has already been ran through GraXpert for gradient correction.

Thrilled to see what you all can come up with!
Like
kgav8r 0.00
...
· 
·  1 like
·  Share link
I do not see any Bias or "dark-flat" frames in this collection.  Do you not have those?
Like
FabianButkovich 0.90
Topic starter
...
· 
·  Share link
I do not see any Bias or "dark-flat" frames in this collection.  Do you not have those?

I do actually let me link them
Like
kgav8r 0.00
...
· 
·  2 likes
·  Share link
Here is what I was able to achieve with your data.  The background was pretty rough.  I noticed that your flat exposure time is listed as 0.00s.  That seems odd.  You might check that out.  Let me know if you want any details, but it was pretty standard Pixinsight processing.

Like
FabianButkovich 0.90
Topic starter
...
· 
·  Share link
Here is what I was able to achieve with your data.  The background was pretty rough.  I noticed that your flat exposure time is listed as 0.00s.  That seems odd.  You might check that out.  Let me know if you want any details, but it was pretty standard PixInsight processing.


Yeah I was aware of the terrible background. Regardless, your edit is honestly amazing! I do want to know details please. The galaxy in your version "glows" and is alot more visually captivating than in my own version. Did you use any sort of masked stretch or did you stretch the galaxy at the same time as the background? 

I think in my own version there is too much contrast between the objects with too sharp of a transition between background and galaxy, which makes it look "flat".

Thanks for sharing!
Like
kgav8r 0.00
...
· 
·  Share link
Fabian Butkovich:
Here is what I was able to achieve with your data.  The background was pretty rough.  I noticed that your flat exposure time is listed as 0.00s.  That seems odd.  You might check that out.  Let me know if you want any details, but it was pretty standard PixInsight processing.


Yeah I was aware of the terrible background. Regardless, your edit is honestly amazing! I do want to know details please. The galaxy in your version "glows" and is alot more visually captivating than in my own version. Did you use any sort of masked stretch or did you stretch the galaxy at the same time as the background? 

I think in my own version there is too much contrast between the objects with too sharp of a transition between background and galaxy, which makes it look "flat".

Thanks for sharing!

Thanks! 

The gradient was pretty challenging which may have been partly due to so many sessions stacked together. The flats may not have perfectly matched each of the lights. Hard to say without having captured it. I used Multiscale Gradient correction at the 512 scale with smoothing at 10. This made a pretty acceptable gradient correction. I try to stay away from GraXpert unless I cannot get acceptable results otherwise. Then I ran SPCC followed by BlurXterminator and then StarXterminator. 

Stretching was done first with Seti Astro Statistical Stretch at .01 median value, then GHS in color mode. It really didn’t take much of a stretch. I followed this up with NoiseXterminator. 

The next step was to take care of the background. This required some creative masking. I started with ACDNR to capture the fine details of the arms. This also captured some undesirable background information, so I isolated the galaxy with the Seti Astro FAME tool. There is still a bit of background captured and you can see the slightly odd shape of the galaxy. I applied this mask to the galaxy and then used pixel math to fill the background with a constant value of .26. Purists may not like this method as it is destructive to the data, but oh well, it give a nice result. 

Now I could focus on the galaxy without worrying about noise in the background.  I used some HDRMT to bring out some core and arm details. Then I used Selective Color Correction in Jurgen’s Toolbox Scripts. I enhance the magenta, yellow, and blue. Then I did an overall Luminance Enhancement which is probably what gives the glow. 

I used GHS for the stars as well in color mode but the arcsinh transformation. This gives pretty good color without making them too bright. I can then screen them in to the galaxy image. A curves adjustment is next to bring down the background closer to .15 so it’s not as grey. The final step is star reduction but that tends to dim them a bit, so I apply the stars image as a mask and brighten them up and add some saturation with the curves tool. 

That pretty much wraps it up and I can save the image and share!
Like
wimvb 3.11
...
· 
·  Share link
Here is the link to all of my data, including flat calibration frames:


Thrilled to see what you all can come up with!

Do you have an integrated image without any further processing done (no gradient correction and no denoising, just straight from the preprocessing step)?
The raw files will just take too long to pre-process.

cheers,

Wim
Like
kgav8r 0.00
...
· 
·  1 like
·  Share link
Wim van Berlo:
Here is the link to all of my data, including flat calibration frames:


Thrilled to see what you all can come up with!

Do you have an integrated image without any further processing done (no gradient correction and no denoising, just straight from the preprocessing step)?
The raw files will just take too long to pre-process.

cheers,

Wim

With OP’s permission, I can post the raw stack I produced.
Like
wimvb 3.11
...
· 
·  2 likes
·  Share link
Here's my attempt on your image, using PixInsight.
The stars are warped; how was the polar alignment and guiding during your sessions?
Process:
  • DBE
  • BXT star correction
  • Classical colour calibration
  • BlurXTerminator deconvolution
  • Arcsinh stretch
  • Star removal (StarXTerminator)
  • Further stretching
  • Detail enhancement using MMT with a mask
  • Colour saturation of  ngc 6946
  • Star insertion

NGC6946_092024_22h.jpg
Cheers,

Wim
Like
blastrophoto 0.00
...
· 
·  Share link
Fireworks.jpg

Here's what I got. Crushed the blacks a bit to accommodate for the gradients. BlurX'd all the things to help recover some detail and to fix star shapes. Seems like maybe guiding was an issue here? Clear skies.
Like
wimvb 3.11
...
· 
·  Share link
he problem with image processing is that it's difficult to stop. Small enhancements, mainly to the core.
blend.jpg
Edited ...
Like
 
Register or login to create to post a reply.