Twice Cooked Lobster (now updated, finished at last!)

[Andy01]

Folks, this is a completely reprocessed and reworked version of the Lobster Nebula NGC 6557 in NB.

Total of 21 Hrs data gathered from the light polluted backyard back in April.

It's been reprocessed in STARTOOLS, which I have really enjoyed learning and using this week - (as you do when it's cloudy and you're stuck in bed with the 'flu).

With Strongman Mike and others deservedly constructive past criticism about my "Paint by Numbers" processing habits ringing in my ears, I'm hoping that I may have had a watershed moment in reprocessing data now.

High res version is here... seems much clearer than my previous attempts.

http://www.astrobin.com/full/169902/C/ (latest version)

I have been gobsmakked by this app. (now that I've taken the time to try learning it properly!) It's an easier learning curve for visual people like me - but backed up by some incredibly intuitive advanced math algorithms.

The Autodev, LP Removal (wipe), Deconvolution, HDR (steady on), Wavelet sharpening and Noise reduction modules work brilliantly but yes, there's more. Star masking, Star reduction and even more features that I havn't properly explored yet, and it works on my mac :D Be warned, it's a memory hungry beast though!

Images were captured and stacked in Nebulosity, processed in Startools and assembled in HST NB Palette in PS.

Highly recommended and at A$60 for the full licence - great value ;)

Comments and feedback welcome - (I have thick skin and it's the only way I learn) :thumbsup:

Final version here - http://www.astrobin.com/full/169902/E/?real=&mod=

[Atmos]

Is it just me or does the hi-res version look like it has been very JPEG compressed? Looks pastel, a bit like an airbrush.

Unzoomed it looks like there is some wonderful detail hidden in there though.

[strongmanmike]

It's a magnificent object this one and the colours are quite attractive but sorry Andy, it looks very posterised :(: https://en.wikipedia.org/wiki/Posterization gives it a water colour painting look.

I'm thinking maybe too many Startools filters this time..? Not sure how to fix it but another shot at it might find the culprit steps in the processing :thumbsup:

Mike

[Andy01]

Thanks for the feedback Atmos, have a look at the Luminance layer here http://www.astrobin.com/full/208173/0/



Hmmm, er thanks he said - (runs off to get strong drink) - gulp, wasn't expecting that! - Did you compare it to the original one? http://www.astrobin.com/full/169902/0/

This was the new Luminance layer after Startools reprocessing - I thought this was ok. I'm unable to make it any sharper than this with my present skill set, but open to suggestions to remedy. http://www.astrobin.com/full/208173/0/

[Atmos]

The luminance layer is looking good, doesn't have the same pastel look, actually has a considerable amount of detail :D

[Somnium]

looking good Andy, i wonder if you can tease out the dust detail in the surrounds a bit more, seems like there is a bit of data there

[Andy01]

Hmmm - ok, thanks for the heads up Colin - stay tuned, see revision below :)



Thanks Somnium, based on that feedback - try try again! Here's a revision backing off one step ie: no Startools star size reduction, and then adding two highpass sharpening layers, one for for small and one for large scale detail at 50% opacity each.

http://www.astrobin.com/169902/C/

[Stevec35]

Looks good on my iPad Andy

Steve

[RickS]

I like the colours, Andy. At full res it looks a little too smooth for my taste.

Cheers,
Rick.

[Andy01]

Thanks Rick, can I please ask you to have a look at version C and let me know your thoughts? Cheers Andy



Chers Steve :)

[Bassnut]

This a bit of a head spin, dont know what to make of this, paint by numbers springs to mind :P. Look, Im a ST fan boy, use it all the time, it has features no one else can match. I dont even know why the hell you would bother with anything else actually. But, jeez, a slider has been pushed too far somewhere here. The power ST has needs to be used carefully sometimes, specially ending the tracking process, Ive had similar problems. The latest version tries to mitigate the likely hood of this happening ive noticed.

[gregbradley]

Overall a pretty good image but as others have mentioned it looks overprocessed. It looks like some of those overprocessed PI images.
I am not against PI but it was used by some to create this artificial exaggeration of existing more subtle details and tonal variations much like narrowband brought out structure not seen in full spectrum images.

But I am a fan of your high impact images. I am sure its recoverable but you probably have to restart early on to regain control.

I think its worth it as this object shows very well in NB as in full spectrum it tends to be another boring all red neb.

Greg.

[Slawomir]

Hi Andy,

I like the second version the most. Echoing others, data looks over processed - I would love to see the result of just stacking, stretching and colour tweaking of this data, without star reduction, wavelet sharpening, noise reduction...simplicity can help to reveal the true nature of things :)

[Andy01]

Thankyou for your incisive comments - Please review this gents...

http://www.astrobin.com/full/208173/B/?real=&mod=

5 steps on the Ha stack - stretch, contrast, decon, sharpen, NR - I would be interested to know where you think the overprocessing look begins.

Warmly

Andy

[Slawomir]

Hi Andy,

I think the last step-NR is too intensive. But in the end, it is your photograph and your workflow, so as long as you like it, that's all that really matters :)

[alistairsam]

Andy,

I think its in step 2, the contrast.
if you look at the first one, the nebulosity seems continuous for want of a better word, as you've increased the contrast, it seems to be broken up.
just using layman terms here.
The contrast module in ST is very powerful and needs to be handled gently.
I tend to keep strength to 45% and dark anomaly to 3, expose dark areas, yes.
Same with the sharpen module. needs to be light to maintain that fluid look.
That's my 2c

Cheers
Alistair

[alistairsam]

Hard to make out with the small snippet, but this is what I was referring to. first to last.

Edit: this is highly subjective and I recall Ivo's version of Eta wasn't widely accepted so its up to personal taste and user perception to a certain extent.
you can sharpen small features without changing the overall look.

[Andy01]

Thanks Slaw, it's all about learning the rules so I can maybe break them later, but knowing how to control which ones I want to break!

See http://www.astrobin.com/full/208173/C/?real=&mod= (New Luminance layer - less NR, more decon see below)



Thanks Al - for this version http://www.astrobin.com/full/208173/C/?real=&mod= I have rejected 4 hrs of data from night two which on closer inspection, revealed a bit of passing light cloud/haze. It's likely that this contributed to overall softness. That said I have eased up on the final NR from 4.5 to 2.5 but BOOSTED the Decon from 1.5 to 2.0 which has sharpened things up considerably, but is now borderline on decon rings. I'll keep an eye on the contrast in future as well :thumbsup:

Here's a draft colour version with the new Luminance added - colours still need a tweak.

http://www.astrobin.com/169902/D/

[Atmos]

It is looking considerably better, there is a lot of detail in the core there that's for sure.

[Slawomir]

I agree with Colin, the new version looks much better :thumbsup:

[Andy01]

Thanks Colin :)



Cheers Slaw - Final version below - I don't think I can improve it any further!

Thanks everyone, for your patience and input - it's been in interesting journey but well worth it :)

http://www.astrobin.com/full/169902/E/?real=&mod=

[Slawomir]

I think the final revision looks awesome!

[multiweb]

Nailed it! :thumbsup: Repro's right on the money, Andy. Superb.

[Andy01]

Thanks Slaw, appreciate your help and input too :thumbsup:



Cheers Marc and thx for your help with this one, I've learned a great deal from this reprocessing exercise, now it's time to move on to something new :)

[gregbradley]

That looks very good. You got a tremendous amount of detail out of this nebula which traditionally in the visible spectrum, is usually a very boring nebula.

Greg.

[plantnerd]

Spectacular and very hubble like

[Andy01]

Thanks Greg -I learned a lot of useful processing procedures revisiting this data so hopefully some of those will stick in my future imaging workflow. Thx for your input, glad you like it now :)



Thanks very much, took a while but perservence seems to have paid off, cheers!

[Paul Haese]

I obviously like the colour palette employed. After all I used a similar palette when I did imaged this object a year or so ago.

You might want to crop the left hand side as there is a stacking artefact running down the panel.

The latest version is still quite noisy and probably needs more integration time to shine or perhaps do a reveal mask to control the noise. I found this particular object to be very tough with regard to noise and it took a lot of integration time to get it to a manageable level and even then I felt as though I could do a lot more to clean it right up. Always a balancing act about perfectionism and rationality when it comes to how much integration time should be collected.

Overall a good image in my opinion, but could do with a bit more.