The Bubble. An ongoing project.

[MattJenko]

I have decided to take a close look at the Bubble Nebula. I have a secluded spot near the house which has a decent view of North, so using that area for now. I plan on updating this thread as more data gets added, so I can track my progress, as I intend to get as much data as I possibly can on this over the next few months, or until it drives me mad. 

First session was one of those where everything was a struggle. Guiding, focusing, everything and then just as I finally managed my first sub, clouds rolled in. Couple of nights later, I managed to get setup in twilight and despite a disaster of a meridian flip, managed a good solid set of 20 min Ha subs.

So here it is to date. 17 x 20 min Ha.  Planning on a lot more Ha and a whole host of LRGB as well, as I am thinking natural colour image, but who knows where this will end up.

RC6, Atik 414ex, HEQ5. Chroma 3nm Ha.

Comments and suggestions welcome.

 

[LightBucket]

Very nice, but me thinks the focus is a tad soft..?

[triton1]

That’s looking good one of the largest bubbles I’ve seen,great detail can’t wait for the end result?

[newbie alert]

Fantastic..

[Stub Mandrel]

1 hour ago, LightBucket said:

Very nice, but me thinks the focus is a tad soft..?

I think focus is pot on, but a quick play suggests that with some gentle sharpening of the original it could be outstanding.

[MattJenko]

10 hours ago, LightBucket said:

Very nice, but me thinks the focus is a tad soft..?

Its probably more my slapshot processing as opposed to focus (I certainly hope so). I have had difficulty getting autofocus routines to work with this setup, so eventually I used a Bhatinov Mask. Given I am going to try and get lots of data, there won't be much need for non manual focusing anyway.

9 hours ago, Stub Mandrel said:

I think focus is pot on, but a quick play suggests that with some gentle sharpening of the original it could be outstanding.

Thanks - one of the hopes of this thread is for me to pick up some pointers on the processing side, as I an RUSTY! I currently only use PixInsight, but want to know if I am missing a trick with my processing here, as I think I am in a bit of a rut when it comes to PI. I have Warrens book and should probably explore more and this project should give me ample practice. As a wise man said perfection only comes from perfect practice, and I am far from that.

[Stub Mandrel]

This with a VERY gentle application of unsharp mask. careful use of wavelets or deconvolution could do better.

 

[MattJenko]

Thanks Neil. I admit to being a little dependent on HDR and LHE in PixInsight. Must broaden my horizons for this I feel.

[Stub Mandrel]

A free trial of Astra Image is good place to start.

[ollypenrice]

I would make a copy layer in Ps, exclude the stars, and run Unshaprp Mask on the bottom layer. I'd then make the top layer active and erase the not-sharpened version with a small eraser on only those features which benefit from the USM. To do this in PI you need to create masks which need to be in exactly the right places. In Ps you can do it using your eyes in real time. To me it's a no-brainer.

Olly

 

[gorann]

Great start. I also thought it looked a bit soft. Could it be seeing? With some almost not noticeable mist in the air I get soft images like that and then suddenly one night everything is very sharp. Maybe you should go for Lum and Ha on the clearest nights?

[MattJenko]

Thanks - I had a feeling I would need to dust off some PS skills, or more likely learn new ones

It does look a bit soft and I think my guiding for some of the subs is not optimal either which might have contributed. This appears to be when the scope was pointing straight up, so think my little HEQ5 wobbled a bit between gear cogs for about an hour or so. Hopefully more data and being more choosey with the stacked subs (plus better processing) will crispen it all up! Not going to get a chance for a few days, so will ponder my shortcomings...

[ollypenrice]

It does look a little smoothed to me, as if given some NR. The excellent structures within the Bubble itself make me hesitate to call it soft, though. Did you run some NR on it?

Olly

 

[MattJenko]

There is noise reduction, a TGVDenoise, although I did Lum mask it beforehand, as LHE is pretty bad for noise, even masked and needed something afterwards. I used only a bias for these as well in terms of calibration, so a little noisy to start with. I think I need to experiment with my workflows and it does seem that a switch to PS for a couple of actions might benefit me too.

[MattJenko]

I have PixInsight and Photoshop. I fear another imaging program may cause my consciousness to implode! 

Here is a Photoshop version. Wow am I rusty with this. I managed another 4 hours (12 subs) last week, of dubious worth but data is data for me at the moment. I think the bright bits of nebula are coming on at the expense of the not so bright. In what was a surprise, it is easier to minimize the impact of processing on stars in PS. Maybe a dual PS/PI future does await me...

 

[MattJenko]

OK. Some RGB to add to this. Had some pretty terrible artifacts in a couple of the channels, so cropped that out for now. Might have to redo them as they are big...

Here is a first effort (first that I am going to post ) which uses the RGB with Ha added to the R.

Feedback welcome. I have better data than I am showing here I feel, and having seen some of the lovely crisp details achieved by some photoshop processors on here, I am not sure my PI approach is cutting it at the moment...

 

[MattJenko]

Slowly but surely. Some deconvolution at the start brought out more detail and I am slowly getting the rest of the processing together. Stars are a bit meh in this version though. II think I need to tone down the range of the nebula a touch and to crispen up the stars, or at least, not have them messed up, which doing the LRGB has done.

 

[Sp@ce_d]

Looks like you're getting there. I've had very similar experience on my Bubble so far too.. just doesn't look sharp enough. I used a very similar setup aswell. RC8 314L+ & 5nm. I've only used PI on it so far.. following Warren's NB processing workflow. I do feel drawn towards delving into PS though. I seem to almost get near & then its just overcooked. Your Bubble itself is showing some good detail. I'm struggling to pull much out of mine.

[MattJenko]

I agree - pulling out sharp data from an oversampled image is tough. I am running at native focal length, a mighty F9, so long exposure times too, and haphazard guiding, so a softness is to be expected at < 1"/pixel I guess, but I still think I can do better. The stars in this are still a bit sucky, despite a better effort at the nebula itself. The Bubble looks nice, and I think it could be sharpened in a non-plastic way a bit further, but the stars could be significantly better. The Ha gives such great detail, but mixing that with RGB with stars about x10 the size is the challenge next methinks as there are halos and a lot of the star fuzziness comes from trying to fix that.

[tooth_dr]

You really have made leaps and bounds with the image.  The last one is really good imo.  I love all the fine detail in the clouds surrounding the bubble itself, superb.

[Shibby]

Getting better and better! I don't think focus is a problem - we're actually looking a quite a narrow FOV here, so you shouldn't expect it to be pin sharp. I like your star colours in the last image very much!

It's a long time now since I've done much processing, but here's my thoughts FWIW. Looking at the original, I would say there's a bit too much "dynamic" contrast, meaning you lose overall range between the bright and less-bright parts of the nebula, looking a bit unnatural. Your second (Photoshop) process is an improvement imho, but perhaps something in-between would be good. Using Ha as full luminance tends to give you that pinkish look to the nebula, so I'd suggest just a hint of luminance (depending on how good the LRGB data is) and the rest as a "lighten" layer or similar to the red channel.

Very nice image, though, and I like your dedication to the target.

[Adam J]

Focus is good here, diffraction spikes are a great indicator of poor focus in a Newtonian, they separate into double spikes when its not spot on. I think the big stars are down to the long focal length and poor seeing.

[MattJenko]

Here is another version which takes the stars from the RGB and is more of a superposition rather than an LRGB combination. It allowed a bit more sharpness without crunching the nebula too much and also makes the image a bit brighter as a result as the stars are less soft. Still not quite right, but slow and steady and all that. Thanks for all the feedback, it really is appreciated. Astronomy and imaging in particular can be a bit of a lonely furrow...

 

 

EDIT: (may be a bit too bright now i see it online!)

[Stub Mandrel]

50 minutes ago, MattJenko said:

EDIT: (may be a bit too bright now i see it online!)

Looks good from here.