My dark nebula is too dark

[Stefan B]

Sorry for the stupid subject title but it's pretty on point... We recently had some partly clear skies and I am worrying that I wasted them. I spent five nights on the Helping Hand Nebula in Cas. It's still a bit early and you don't even need a meridian flip at the moment. But I was looking forward to imaging the target and so just gave it a shot. 17 hrs of data were gathered over five nights. Quality is probably a bit mixed (high clouds etc.) but should contain some decent data.

I stacked the first time after two nights and six hours of data and immediately had the impression that this might be more challenging than I anticipated. Gosh, is that thing faint. So I just added more data. This is what I settled upon after several processing runs:

This is kind of a middle ground of my attempts. Somewhere between the one where the noise is screaming at you and the one where you need dark adapted eyes and averted vision in order to see the nebula on screen.

See https://www.astrobin.com/hn4qew/ for technical details. On Astrobin it's still in the staging area.

By binning 50%, ignoring fine detail, adjusting ROI, reducing shadow linearity, applying DimSmall and Shrink I tried to push back the noise and highlight the nebula. It was quite hard to find the right balance. And still the nebula itself is pretty noisy. (Disclaimer: In order to highlight the nebula a bit more I used the 'Equalize' option in PS but only with 4% fill opacity.)

My question basically boils down to: How to process dark nebulae in StarTools?

Here's an image that was acquired with VERY similar equipment and only marginally more data: https://www.astrobin.com/eojxrd/ The location is described as "my darkest place in Austria" so probably better than my Bortle 4/4.5 balcony. Still the difference to my image is insane. Is this only due to star removal, stretching as hell and denoising as crazy before adding stars back. Or is there more to it?

If anybody wants to give it a try, here's the ASTAP stack:

https://www.dropbox.com/scl/fi/iihad428 ... 07yoq&dl=0

I struggled with Wipe. An unusually high value of gradient aggressiveness (>90%) and gradient falloff (70 - 80%) was necessary for an even field. One side looks a bit like a flat problem. But flats were new and done with NINAs flat wizard. I always had good results by using this approach. For 17hrs of data there's also an surprisingly high amount of correlated noise. I used a correlation filter of 0.5 pixels.

Feedback welcome!

Regards
Stefan

[Mike in Rancho]

See also, my shower is too wet and my oven too hot.

Good try though Stefan!

Using my finely-honed Google skills, it seems the little valley that IOTD was in, Hinterwildalpen, is B2 and maybe also mountain-protected when aiming at Cygnus. A little bit higher perhaps than Feuchtwangen but not majorly so not sure altitude is an issue. Unless he's up a hillside. Also maybe seeing and transparency, you just never know.

I'd have to find my Bortle converter, but with everything else so very similar as to time, camera, and scope, that's probably going to make a decent difference in the achieved SNR. More time might allow you to bring out more.

I'll give the file a try (if I can remember how to process!), but at first glance that's what it seems like to me. Similar to when you guys can put up clean images containing dust that has a nice orange-brown color to it, but my own looks like gray lumps of coal.

[Carles]

Hi Stefan, nice image

But not as good dataset, mainly because of ASTAP artifacts.

I struggled with Wipe. An unusually high value of gradient aggressiveness (>90%) and gradient falloff (70 - 80%) was necessary for an even field. One side looks a bit like a flat problem. But flats were new and done with NINAs flat wizard. I always had good results by using this approach. For 17hrs of data there's also an surprisingly high amount of correlated noise. I used a correlation filter of 0.5 pixels.

I'm sure they are giving you issues with Wipe and overall processing and colour.

The artifacts are from dithering , for some reason sometimes ASTAP doesn't reject them well, i've talked about it on my comparision long time ago, and hence I stack with Siril.

Here's what I mean, and they're all throughout the image, the red, green and blue "blobs"

ASTAP artifact.png (422.46 KiB) Viewed 1360 times

Nevertheless, I tried correcting background with GraXpert then basic Startools processing (just autodev, equalize on Contrast, some HDR and Superstructure. ) and got this. (did bin 50% just to speed up)

Helping reduced.png (419.23 KiB) Viewed 1360 times

Maybe you can figure out how to setup ASTAP to reject properly those hot pixels, otherwise, try stacking with other app.

Regards,

Carles.

[Carles]

Did another test, without pre-proc with GraXpert

Helping reduced no graxp.png (408.51 KiB) Viewed 1359 times

Still OK, but I think the other one is better. the blotches of artifacts really affect.

Cs,

Carles.

[Mike in Rancho]

Good test, Carles.

I noticed those also. At first I was thinking those waffle artifacts could be something from debayering, and ASTAP also has a lot of extra options regarding that and color - though I would expect Stefan to keep to the "approved" settings.

Once I got to color and noticed the blue green red splotching it started pointing towards rejection. Still strange that they are so waffled rather than the streaks we usually see.

Maybe even just tightening the KS limits would help here.

But indeed I wondered how those artifacts, being read as signal, would bother Wipe and the rest of processing.

During the Wipe courtesy superstretch, the hues of the dark nebula were indeed more like a rich golden brown. But I couldn't maintain that when I got to color. Heavy saturation just blew everything out, and otherwise I ended up with lumps of gray charcoal.

Of course I could be out of practice too.

[decay]

I too would assume, that something went wrong with stacking:

2024-09-03 18_52_37-StarTools.jpg (87.98 KiB) Viewed 1331 times

Never seen such artifacts before. As for the gradients: I also encountered such problems before. These gradients do not look very special, but it is difficult to get rid of them using StarTools' Wipe module. As you said, only by using a high Aggressiveness setting or even by using the 'Uncalib1' preset. Probably it is better to use some other external tool beforehand, like Carles suggested.

Good luck!

Dietmar.

[Stefan B]

Thanks guys for taking such a thourough look. I noticed some of the artifacts but just saw how abundant they are after you pointed at it. That could certainly throw off Wipe.

I actually tried Graxpert on my first stack which contained only six hours. It wasn't much better than the Wipe result so I didn't bother trying again on the 17 hours stack.

It reminds me of my NGC 1333 data set which had similar problems if I remember correctly. My results with Mike's Siril and PI stacks were much better. I will give Siril a try. Somehow last time I didn't get it running. Which is pretty embarrassing since obviously lots of people are able to use it...

It will take me a while until I'll be able to test Siril. But I'll keep you updated

Regards
Stefan

[Stefan B]

Okay, couldn't wait for giving it a try. Stacking with Siril and scripts worked like a charm. No idea what I have done wrong last time. Maybe my folders contained files that didn't work so it constantly gave me error messages. Anyway, it worked.

@Carles thanks so much for pointing me to another stacking software. Revealing the dust was much easier now.

This is what I got after a quick processing run:

Actually I am pretty happy with that.

If anyone wants to play around with the Siril stack, there you go:

https://www.dropbox.com/scl/fi/iihad428 ... jwn5o&dl=0

Thanks again, guys! That saved my image and 17 hrs of clear sky time

Best regards
Stefan

[Carles]

Hi Stefan!

That's a great result! Glad it helped .

If you feel there's lack of signal or contrast (comparing to the other guy at astrobin ) I'd suggest analysing the subs and see specially hfr and star count( i find it easier to do this in DSS ). Could be some dog or thin cloud in some subs that might skew the stack. You might end up with less total integration time but better signal to noise ratio. (Unless you have already done that! Then ignore )

I'll give a go to this new stack when I get home, it's an object i got on my list too, will be a good practice!

Cs,

Carles

[Carles]

@Stefan B I think you've uploaded the ASTAP stack again, as I just downloaded it and it shows the artifacts still.

Carles.