[RCC] DSO processing workflow - am I going in the right direction? (here M42)

13 replies423 views
dzambon avatar
Hi everyone,

I am quite new to AP and I've been trying to improve my processing technique.
  • Q1: Does anyone of you have some feedback to share with me? like, about the overall look or specific aspect I should take care of?
  • Q2: I've tried to follow several suggested procedures, like those in the channels Cuiv the lazy geek, Nebula Photos, Astrobackyard, AstroFarsography... However I couldn't replicate as good results as theirs on my data. Could it be too specific to the equipment and actual collected data? Anyways, I borrowed ideas from all of them and tried to come up with something good for me. Should I try to stick to more "consolidated" workflows or work on my own?


Here is the picture I've got from a 55x120" integration taken with an unmodded Canon 1200D, Tamron tele at 300mm f5.6, StarAdventurer.


Great Orion Nebula (M42)



In my mind I wanted to:
  1. 1. maintain a sort of natural look
  2. 2. make visible all the nebulas details I've collected
  3. 3. keep the processing workflow as simple as possible


More in details, my main concern with respect to the previous version of the picture was to
a. keep the stars decent and "homogenous" over the entire image, even those overlapping with the nebulosity (and given the aberrations of my Tamron lens)
b. make a bit more visible the dark strands of dust
c. not play around to much with manual masks to let some specific parts stand out more

The workflow was the following:

__Siril__:
  • - crop and background extraction
  • - rough color calibration, and conservative histogram stretch
  • - contrast-limited adaptive histogram equalization (grid < 3 and threshold ~3 to preserve stars)
  • - split into separate R, G and B channels


__Gimp (channel-wise)__:
  • - create a synthetic luminance from RGB
  • - G'mic Iain's noise reduction for the color channels to remove future saturation artifacts
  • - Just a little bit of G'mic Iain's noise reduction to the luminance channel on the shadows
  • - recombine RGB


__Gimp L+RGB__:
  • - Keep separate layers for luminance and RGB
  • - adjust color calibration and saturation of the RGB layer
  • - play with curves and G'mic local contrast enhancement on the luminance
  • - merge lum and RGB
  • - star reduction with value propagate
  • - final small curves and noise adjustment

  • Q3: I've seen I was able to pump up the nebulosity a lot with Starnet, however I've experienced several artifacts associated with the aberrations of my lens. Would you suggest to include it in my "standard" workflow?


Thanks a lot!
Daniele
Helpful Engaging
Alberto Ibañez avatar
Hi,

I can't guide you about the tools because I don't use Siril/Gimp for processing but I can enumerate some aspect that you may want to improve.

-The star shapes are not round and the overall image is a bit blurry. Seems like the focus is off.

​​​​​​-The color balance seems a little off too. look at the brightest blue stars, they are a bit green near the core (for my taste).

-The core is clipped. You probably need to take shorten exposures.

-The background is too dark. Again, this is a personal preference. I prefer to keep the "true black" of the image at a brightness not less that 0.1 (for a 0 to 1 range).

-The transition from the background to the dimmest nebula features has "holes". This can happen if you pushed up the brightness using a mask that is too noisy. Don't know if it is the case…

Hope it helps! smile
Helpful
Olaf Fritsche avatar
It's a very nice picture! Be proud of it! 

M42 is easy to see, but not easy to photograph. There are so many structures with different brightnesses that not all of them go on one picture with one exposure time. 

My tip for this object: also take photos with shorter exposure times and combine the areas with the appropriate exposure. This way you can resolve the outburned area in the heart of the nebula. 

When processing the image, you can try to increase the contrast with a high-pass filter and sharpen the image this way.

But don't be too critical of your own work! Show it to your friends and family, and you'll reap an amazed Wow!
Well Written Helpful Insightful Respectful Engaging Supportive
Jared Holloway avatar
Everything Alberto says is spot on - wanted to add (and it is something I am still working on) - is that you can take multiple exposure sets, for example: 40 x 120s and then do 20 x 10s and blend the two together, especially at the core of m42, to get a less blown out look. The issue I have personally is that all my longer exposures of Orion are too long to get a good blend. I am not sure how to do that in Gimp, but if it works similarly to Photoshop it can be done.

Beyond the over-blown core, the main issue I see is the focus. Everything else can be fixed with processing.
Helpful
Jared Holloway avatar
Olaf Fritsche:
It's a very nice picture! Be proud of it! 

M42 is easy to see, but not easy to photograph. There are so many structures with different brightnesses that not all of them go on one picture with one exposure time.

This is the number on thing to keep in mind with m42! Very hard to get that "perfect" result!
dzambon avatar
Great, it seems I have material to work on! I appreciate all you comments

Focus: That's an interesting comment. Probably I have been convincing myself to have focused correctly from what I was seen in 3" display of the DSLR, and always blamed the lens quality
I will do more tests to see how much I can improve in this respect

Saturated core: That's true... I should have cared more. Noted!

Star shape and color: Perhaps, the asymmetry in the star shapes is due to the extensible part of the zoom lens which is a bit wobbly... that's my guess.
Moreover, the same star has a different shape in each channel... not sure how to fix it, apart from upgrading the gear

Background, sharpness and holes: I see what you mean, I'll try to improve over them.
About the holes, actually there are no masks involved. I'd say it's the combination denoise + local contrast pushed to much...

Thank you all! also for the encouraging words, Olaf and Jared
Daniele
Alberto Ibañez avatar
Hi, 

For focus, I suggest you to use a bahtinov mask. Works great and it's very easy to make it with a piece of black cardboard.

About the holes, your statement makes sense… maybe this is a part of the workflow to be refined…

Good luck and show us the results! smile
dzambon avatar
Alberto Ibañez:
For focus, I suggest you to use a bahtinov mask. Works great and it's very easy to make it with a piece of black cardboard.

The point is that I am already using one

Anyways, last night I made a third attempt to keep the processing defects and artifacts under control.
I kind of set this @Jared Holloway 's picture as target, which I like

- I made sure the background was not clipped and made it a little bit brighter (gimp says median is around 1% of the range, but at 10% of the "perceptual space", not sure that is what you meant Alberto)
- I set a white reference to fix the previous cyan halos around the brighter stars, and the overall tint has shifted quite a bit. But there are still some grayish halos that I wasn't able to deal with
- not sure it matters with the sort of "washed out look", but I haven't mentioned that it's a 50% crop in height and width of the original pic.

Thanks again!

Jared Holloway avatar
Anyways, last night I made a third attempt to keep the processing defects and artifacts under control.
I kind of set this @Jared Holloway 's picture as target, which I like


Thank you! This process looks good (in my opinion!) - m42, as I mentioned before, is still a work in progress for me - mainly for the core and its brightness. I think with m42, in many respects its personal preference on how bright / blown out you want the core - some people like it not bright at all, some like it bright. I am an in between person, I like it sort of bright but with still being able to see the Trapezium.
Olaf Fritsche avatar
Jared Holloway:
Anyways, last night I made a third attempt to keep the processing defects and artifacts under control.
I kind of set this @Jared Holloway 's picture as target, which I like


Thank you! This process looks good (in my opinion!) - m42, as I mentioned before, is still a work in progress for me - mainly for the core and its brightness. I think with m42, in many respects its personal preference on how bright / blown out you want the core - some people like it not bright at all, some like it bright. I am an in between person, I like it sort of bright but with still being able to see the Trapezium.

For the trapezium you need a very short exposure time. So M42 is a blend, no single malt. ;-)
Well Written Insightful Respectful Engaging
Jared Holloway avatar
Olaf Fritsche:
For the trapezium you need a very short exposure time. So M42 is a blend, no single malt. ;-)


Yep! My struggle has been the blending personally... for the two data sets I do have to work with I think my normal exposure was a little bit too long. One was a wide field that also featured Horsehead and m78 and I wanted to pull out the detail in those. I am toying with the idea of having three exposure sets - one for everything but m42, m42 and then a short for the core.
Dale Penkala avatar
Very nice image, be proud of what you do as WE ALL are learning in this whole world of processing images!

I’m far from an expert and pretty much everything mentioned about over exposure of the center/core is the biggest thing I see.  Would definitely shorten exposure time. I always have that problem myself! 

I’m a really big fan of Gimp and I use it for most of my post processing and LOVE the G’mic plugin! Just a couple of things I have found to work well for me in other images is the “JPEG Smooth” filter and there is a wavelets filter that I’ve found to work quite well to help smooth the image before any sharpening that I do. I’ll usually do the Wavelet thing 2 times maybe 3 if it looks like its working. 

As far as sharpening goes, there is another plugin that I found called Py-Astro and its got a few sharpening options that I really like and use over the normal “Unsharp Mask” called “SMART SHARPEN” 

I don’t now if this is a standard practice or not but when ever I am in GIMP and do something specific I like to make a duplicate layer and use the “Opacity” in the layers window. I especially do that for the “Star Reduction” and “Sharpening” that way if I think I may have over done a process I can easily back the process off or tweak the strength of it. Again I really like that for the Star Reduction.

Again I’m not an expert thats for sure but I do love Gimp and found it to be a really solid tool for someone that doesn’t like subscription based softwares.

Keep up the great work!

Dale
Helpful Respectful Supportive
dzambon avatar
Thanks for the pointers @Dale Penkala !
make a duplicate layer and use the “Opacity” in the layers window.

That's what I do myself, as well. I use it also to create checkpoints to compare different versions; I simply never got used to the history window but I am no expert, so there is a good change I am missing the proper way to do it
Dale Penkala avatar
Thanks for the pointers @Dale Penkala !
make a duplicate layer and use the “Opacity” in the layers window.

That's what I do myself, as well. I use it also to create checkpoints to compare different versions; I simply never got used to the history window but I am no expert, so there is a good change I am missing the proper way to do it

Ya I love the “opacity” slider for the strength/variability of the layer.
Once you get used to the “history” window you will like it because you can just go back and undo anything you want without undoing everything before it or what ever you have done prior to sharpening, star reduction, etc…

Dale
Related discussions
[RCC] Soul Nebula: Where is my OIII?
Hello everyone. I recently purchased my first dedicated astronomical camera (Playerone Ares-C) and am finally starting to get some data that can be considered as such. After a few short acquisition tests, I finally managed to acquire a substantial da...
Nov 3, 2024
Both posts are from newcomers to astrophotography seeking advice and feedback on their imaging techniques and results.