1.7.424 alpha with new Wipe module (M31 question)

[fmeireso]

Still confused witht the wipe function.

After wiping, and choose the color button i get still this , see image

Can't understand why.

Found that i had to reduce the gamma button an eventally got a descent image, but is this to be expected, is what i wanne know...

[alacant]

is this to be expected

Hi
My guess is that wipe should be able to flatten the colour better if you cropped the bottom edge a few pixels, but unless you post -a link to- the original file, it is just that; a guess.
Cheers and clear skies,
Steve

[fmeireso]

is this to be expected

Hi
My guess is that wipe should be able to flatten the colour better if you cropped the bottom edge a few pixels, but unless you post -a link to- the original file, it is just that; a guess.
Cheers and clear skies,
Steve

Indeed there were quite some stacking artifacts , still cannot understand why. Cropped it allready quite a bit...



The stack...

https://www.dropbox.com/s/9fceaoz96iw0jwb/M31.TIF?dl=0

Anyhow inthe color mode of autodev, reducing the gamma, it came eventually out pretty well.

However, this did not happen with my NGC 7000 although i used the same stacking parameters, in fact those suggested by ST.

Bit confused, but not dissapointed

[admin]

Thank you for uploading! (I moved this into its own topic)

First off, your dataset appears color balanced. Please turn off color balancing in your stacker (for recommended settings for popular stackers, see here).

Stacking artifacts are indeed visible; zoom into the image and look at the edges. You will find a 1 pixel "border". After running an inital Wipe another 1 pixel column at the left edge can be seen as well. You really need to crop more than that as there are still other artifacts visible along the edges;

Selection_219.jpg (477.39 KiB) Viewed 4087 times

That said, the new Wipe can model and correct such artifacts a lot better now. And giving it a mask that contains the areas it may sample yields an even cleaner result;

StarTools_220.jpg (414.56 KiB) Viewed 4087 times

StarTools_221.jpg (483.51 KiB) Viewed 4087 times

When you "Keep" that result and launch AutoDev, all you need to do is specify an RoI over a slice of the galaxy to give it an idea of the sort of detail it should optimize the stretch for (see here on what AutoDev does and how it works);

StarTools_222.jpg (367.61 KiB) Viewed 4087 times

From there you can enhance local detail, etc.

For coloring, when you hit the Color module (preferably one of the last steps), we can see that the default settings StarTools comes up with are too green (use MaxRGB mode);

StarTools_223.jpg (481.68 KiB) Viewed 4087 times

Once we are absolutely sure the color balance is now correct, we can use the Cap Green function to kill any spurious green dominant pixels.
There is some slight chromatic aberration (or channel misalignment?) visible (also the cause why StarTools suggested a default color balance that was too green). The Highlight repair setting helps correct aberrant coloring in the highlights. See here on how to use the Color module and how to color balance correctly.

You'll end up with something like this for example;

StarTools_224.jpg (298.55 KiB) Viewed 4087 times

A good yellow-ish core (older stars) and bluer outer rim (younger stars) and red/brown dust lanes. Perfect!

Hope this helps!

[fmeireso]

Thanks alot Ivo,

Really appreciate this.

I will recheck me DSS settings, guess i missed background calibration setting i noticed.

Last picture is really great!

Strange i winded up with that many stacking artifacts, allthough this fall beyond the scope of ST.

Some CN members reported also quite some 'walking noise' in the stack...cause of not dithering, i will do this on next occassions..

I will redo my sessions following your guidelines...!

Thanks again!

[fmeireso]

• 

  M31_2.jpg (328.25 KiB) Viewed 4025 times

Guess i got it out..took about an hour with trial and error.

I also reduced the starfield a bit, needed a mask for both M31 and M101, to keep untouched, otherwise i lost too much detail.

Trouble is for very cluttered starfields this seems really difficult. In this case it worked pretty well...

Did a final touch in GIMP only to blacken the background a bit, not much, just a bit ...

[admin]

Guess i got it out..took about an hour with trial and error.

I also reduced the starfield a bit, needed a mask for both M31 and M101, to keep untouched, otherwise i lost too much detail.

Trouble is for very cluttered starfields this seems really difficult. In this case it worked pretty well...

Did a final touch in GIMP only to blacken the background a bit, not much, just a bit ...

Great work!