SH2-155 - The Cave Nebula

[decay]

Hi all,

At the beginning of November there have been a few clear nights here at my location, which I used to gather some data of the Cave Nebula. My intention was to acquire broadband- and NB data in order to combine them using the NB Accent Module. But the quality of the broadband data was unusable, so I decided to process NB data only and to layer in RGB stars of one night:

Please find technical details on AstroBin:
https://www.astrobin.com/rc1d0z/

The processing in ST was as follows: Compose module, open NB data as Bicolor OSC/DSLR (HOO processing). I had no luck using OptiDev in this case. Too much noise was introduced, no matter what I tried. The data is probably too weak. FilmDev gave a more pleasing result. I lowered Gamma to 0.92, which brought out the main arch of the Cave nicely. Nothing special about using the Colour Module – no need to do independent non-linear stretching per channel as there’s almost Ha signal only. In the Superstructure module the ‘Strength’ parameter was reduced to 60%. The resulting image was slightly post-processed with darktable (moderate adjustments of contrast and saturation).

The loss of four nights of gathering broadband data was quite annoying, but in the end I think I got a relatively ‘natural’ looking result.

As always: Comments welcome

Best regards, Dietmar.

[Startrek]

Hi Dietmar,
Nice result and a really natural look too.
Stars have excellent color
Well done !!
I’m still using ST1.8 ( haven’t felt the need to jump up to 1.9 ) for me FilmDev tends to blow out my detail including stars. I agree it seems to work better than AutoDev in regards to noise control but not really.Experimenting with AutoDev with noisy data , I’ve found you have many features and tools to peg back your noise floor. Ignore fine detail , creating an ROI and gamma adjustments work better than the FilmDev slider as you still maintain that fine control of dynamic range and brightness. Gamma in AutoDev is excellent when you use it in combination with Ignore fine detail and an ROI.
I’ll only resort to FilmDev when my data is extremely poor and ridden with noise grain both course and when imaging Broadband targets like faint galaxies under my B8 City Suburban skies.
I know Ivo is not a fan of folk using Film Dev , he will always suggest to try AutoDev ( OptiDev ) first and use FilmDev as an absolute last resort.

Clear Skies
Martin

[decay]

Thank you Martin

Yes, in almost all cases the results of OptiDev/AutoDev are of superior quality compared to FilmDev. But there are some rare cases when FimDev does a better job for me. I don't know why, but I guess, it's the very poor quality / weak signal of my dataset in this case. So yes, it was the last resort this time It's not the noise floor in particular, but the applied stretch as a whole, causing unexpected overall appearance and sometimes noise even in parts with higher SNR.

Well I guess on fine day I will do the jump to mono imaging too. For this target it would have been helpful to have full pixel surface available and not only 25% of it. But, to be honest, I'm too stingy for that at the moment

I’m still using ST1.8 ( haven’t felt the need to jump up to 1.9 )

Absolutely fine. With 1.9 still being beta it may be subject to changes and one will always need some time to get used to it. For me, there are some features that I would not like to miss. Especially SV-Decon often does a better job for me than the old version of 1.8. (I know, this is not true for all of us / for all datasets.)
My other motivation to use the 'bleeding edge' versions is, that it is probably helpful for Ivo to get some feedback to new features or improvements, especially in case of troubles.

Thanks again, Dietmar.

[Stefan B]

Hi Dietmar,

very nice image! If you would have told me that it's pure broadband, I wouldn't have questioned it. I never ended up with such a natural appearance after using duo NB data only

I also think you're broadband stars are remarkable. So good job with the mask! Did you have to fiddle a lot manually? And did you also layer in the color of the brighter stars with spikes? I always found that very challenging and often left out these stars and only changed the color of the smaller stars, which were easier to mask.

Regards
Stefan

[decay]

Hi Stefan,

thanks for your reply

Did you have to fiddle a lot manually? And did you also layer in the color of the brighter stars with spikes?

Oh yes, that's maybe worth mentioning. No, no extensive fiddling. First, I'm a quite lazy guy, at least regarding boring work and second, I've never been good in art class, just like Ron Basically following Guy's great user notes, what I did was running the layer module two times. First run with a default generated star mask (without tracking), which looked like this:

mask1.png (114.38 KiB) Viewed 18180 times

Second run, a painted mask containing circles for the halos of the larger stars:

mask2.png (4.76 KiB) Viewed 18180 times

This time I increased the 'mask fuzz' parameter to 25px. That's all

Best regards, Dietmar.

[Stefan B]

Hi Dietmar,

that's a clever way to do it. Also smart to read the user notes. I'll keep that workflow in mind.

Regards
Stefan