polslinux wrote: ↑Fri Sep 11, 2020 2:39 pm
Hello all,
Hi and welcome to the forums (and the hobby!)
As this was my first try ever, I wanted to keep things simple, so I captured only lights (had to deal with the all polar alignment, focus, etc).
The problem is that not acquiring flats is a false economy; as you're finding out they make your life a lot harder - not easier!
Because you did not calibrate with flats you now have to try to process out gradients and separate the galaxy from the murk. This is more difficult (and will always yield suboptimal results) than just being able to "trust" your data as it came out of the stacker.
Your default workflow is sound though (if your dataset had been calibrated with flats) and you've brought out all the interesting coloring.
Due to the strong gradients and uneven lighting, you will have to help Wipe a bit more with figuring out what is gradient and what is signal.
To do this, first roughly mask out (you can see how embarrassingly rough I drew the shape below
) the area that is definitely signal (or definitely not signal in the case of artefacts) and, hence, should be off-limits for Wipe to look for an interstellar background. I use the lasso tool for this.
-
- StarTools_205.jpg (148.49 KiB) Viewed 3652 times
Then, back in Wipe, I used the Vignetting preset and further increased the Aggressiveness until most of the uneven background disappeared. I'm say "most" here, as Wipe will relentlessly keep showing your how "bad" you background is through its courtesy AutoDev for diagnostics purposes; your are of course not obliged to stretch as hard as the courtesy AutoDev is doing here on purpose;
-
- StarTools_204.jpg (408.03 KiB) Viewed 3652 times
After "keeping" the result, you do an AutoDev once more - this time for your real/final globl stretch using the cleaned up data. As with all images that contain an interesting object on an otherwise uninteresting dark/noisy background, be sure to specify a Region of Interest for AutoDev to optimize for;
-
- StarTools_206.jpg (184.56 KiB) Viewed 3652 times
Do whatever else you fancy doing to the image (with low signal, you will notice that StarTools becomes much more careful with how/where it applies enhancements, as everything is constantly balanced against how badly noise grain will increase - it is much harder to "overcook" images in StarTools because of this).
If you now run the Color module you should see the same roughly the same coloring you already accomplished. However sanity-checking the coloring with the
MaxRGB mode reveals that the galaxy's core is too green-dominant (it should be yellow due to gas depletion leaving only older stars). Keep increasing Green Bias Reduce until the core no longer is. Once you're satisfied your color balance is correct you can decide to set Cap Green to 100% to kill any left over green noise. If you think your data can handle it, you can decide to introduce a little more saturation in the darker parts.
All in all what you want to see is a yellow core, a bluer outer rim, red/brown dust lanes, a good random distribution of foreground star temperatures. E.g. something like this;
-
- StarTools_207.jpg (292.95 KiB) Viewed 3652 times
Turn off Tracking, avail of the targeted data mined noise reduction to your taste, and you'll end up with something like this;
-
- m31_32bit_DSS.jpg (372.45 KiB) Viewed 3652 times
Priority #1 should be flats. From there on it's just a matter of collecting more data and you're on well on your way!
Hope this helps.
