"Eye of God" - NGC7293

[jase]

Greetings All,
We've had a lot of Helix nebula pics contributed to the forum of late, so here is my rendition of NGC 7293 – The Helix Nebula (http://www.cosmicphotos.com/gallery/image.php?fld_image_id=114&fld_album_id=11).

The Helix Nebula (NGC 7293) resides in the constellation Aquarius and is dubbed the “Eye of God”. It is one of the closest of all planetary nebula at a distance of approximately 450 light years and is about half the size of a full moon. Despite its size, the light is spread over a large area making it a difficult object for visual observation. The nebula displays different ionisation levels of ejected matter from the dying central star. The inner blue hue indicates the presence of excited oxygen atoms, while the vibrant outer structure consists of hydrogen and nitrogen atoms.

About the image;
The image is a [LHa]RGB composite with a total exposure time of 4.5 hours (L:60min,Ha:60min,R:50min,B:50min,G :50min). The Luminance and Ha data I combined was really weak/crap. I’m not overly pleased with the resolution obtained. Two hours should have been sufficient to bring out the cometary knots. It also didn’t help that I collected the luminance data when the object was relatively low. May revisit another time to collect better luminance data. My imaging goal for this target was to not just too bring out the show piece object, but also lift the image to display its environment. I kept this in mind with processing the data and made subtle tweaks to emphasis features of interest in particular background galaxies. I actually prefer this target in narrowband compared to RGB as the mapped colours really gives an extra dimension/wow factor. In my opinion, RGB appears a little flat. For the record, colour balance is slightly out. The inner shell should have a greenish tint. I attribute this to either lack of RGB data (likely cause) and/or the Astrodon filters are not true 1:1:1 ratios, but 1:1.2:1.2 for the specific imaging set up. May need to perform some G2V tests to confirm.

Image Processing;
The typical dark/flat/bias calibration. Registered in Registar. RGB was 2x2 bin, thus scaled to luminance (1x1). Median combined in MaximDL. Data was extremely noisy so plenty of image cleaning performed. Hot and dead pixel removal performed using the RC Console in MaximDL. A few major gradients to deal with, one of which took some time to process out. Primarily worked in MaximDL to remove these. Lum and Ha blended in PS using lighten mode. Shadow/Highlights tool used for non linear chrominance stretch. I blended the luminance data in twice to obtain better control over colour saturation (first at 40%, the second at 100%). Minor noise reduction, though think I could have gone harder as the data is still noisy.

Thanks for looking, hope you enjoy.

Cheers :D

[seeker372011]

plenty of background galaxies there! IMHO looks most pleasing at mid size (1600x1200)

[RB]

Your polar alignment is way off Jase ! :P :poke:

J/K

Oh my gosh I can't believe how many background galaxies you've captured and I love the fine wispy outer arm and gorgeous colours.

Stunning work mate.

:eyepop:

[Tamtarn]

That's a wonderful capture Jase

We agree the background galaxies and the crescent wisp makes this a very special image :thumbsup: :thumbsup:

[glenc]

Thanks Jase, another great image. :thumbsup:

[iceman]

Stunning result, Jase! I love the wispy outer shell of the nebula and the inner pink globules where the blue meets the pink.

Incredible detail and colour.

[h0ughy]

wow, simply stupifying!!!! that is one of the best shots of the helix I have seen. good gear and good knowledge and operation give results like this.

[jase]

Thanks Narayan. :thumbsup: Agree the 2400x1800 loses much of the resolution. This is actually a crop of the full frame which is 4008x2672. If I had of resized the original I could have possibly gone with 3000x2400.



:lol: :lol: Perhaps by 10 +/- arcseconds :D. Thanks for the comments Andrew. :thumbsup: It is a pleasure to capture this object at a long focal length.:)



Cheers Barb and David. :) I'd say its a keeper. Also, I look forward to seeing your imaging output on the new mount - congrats.



Thanks Glen. Pleased you like it.:)



Cheers Mike. The wispy outer shell came from the Ha data. Though was hoping I'd get more from this channel. As I mentioned the luminance info was rather weak. Also some dithering issues didn't clean the noise as I would have hoped - though think this is a limitation of the set up considering I was also guiding at the same focal length. Can move the guidestar too far otherwise it will miss the guide chip all together. Thanks again.:thumbsup:



Thanks Houghy.:) Pleased you like it. The version displayed is the fourth rendition of processing. Just wasn't happy with the first three as I tried different blending options of the lum and ha data. Not perfect, but then again what is? Cheers.

Thank you again for all the comments.

[Ric]

Hi Jase, a wonderful image, well done.

There is an amazing amount of fine detail in that image, all the wispy little trails on the inner side of the ring are fantastic and the colours are very natural as well.

I also enjoyed the galaxies that you captured as well, there is a good amount of detail in them as well.

A very fine effort.

[Garyh]

A very nice rendition Jase!! :thumbsup:
very sharp and detailed, I like the colors you have got with the processing. A nice blend of the Ha into the LRGB.
One thing that seems to stand out to me is the central blue halo in the very centre seems to stand out like a ring around the central star. I havn`t really noticed that before in other images.
Verry nice.:)
cheers Gary

[Dr Nick]

Absolutly beautiful! ;)

[skeltz]

Good effort there jase,yes i never knew there were that many galaxy's around the helix.
must be happy with that one !!

[Lee]

Hubble-esque.... awesome.... :eyepop:
Love seeing all the background galaxies.

[Dietmar]

very well done, Jase!

[strongmanmike]

Is that a new RC + PME..? Or have you always had'em? A nice pairing and becoming somewhat common place in OZ of late it seems? :rolleyes:

That's one of those images (like I've been doing a few of lately :lol:) that screems.."Aaarg if only I had more exposure" :doh: At F9 ya gunna need a lot of data on that object that's for sure. All the hallmarks of something special are hinted at and given some more exposure and work on the noise you can see it will produce a stunner!
Yes, there are some galaxies around there huh? :eyepop: You mentioned that you processed to enhance them, how did you do that?

I have decided that the 30% or so QE in the Red of the interline KAI11002 chip, not to mention the terrible noise, is just not good enough so I am about to upgrade to the superior and larger 16803 full frame chip in my ProLine :D. Wolfgang Promper reports that darks are not necessary with the 16803 either.

Nice work Jase

Mike

[danielsun]

WOW Jase!!! :eyepop:That is truly an absolutely fantastic shot, and with the distant background galaxys!!! :thumbsup:

[jase]

Thanks Ric.:) I feel it would have came out better with a clean luminance channel, but the result it pleasing none the less.



Cheers Gary. The ring around the central star came from the luminance data, not the Ha. This is the reason why I blended to the two, before combining it with the RGB data. In doing so, you get the best of both worlds. Thanks again.:thumbsup:



Thank you Dr. Nick :)



Cheers Rob. :) Plenty of background galaxies indeed. I was going to create an auxiliary image that was annotated list the names of each, but haven't had the time. Thanks.


Thanks Lee. Pleased you enjoyed it. :thumbsup:



Cheers Dietz:)



:lol: Nope, the scopes not mine Mike. It's Brad's from Southern Astro. Though having stated this, I'm contemplating getting a 12.5" or 14.5". They are a considerable investment, but the quality speaks for itself. Will probably need to ditch the Losmandy Titan even though it will handle the 12.5" with ease. Anyway, that's another chapter in a long story.

Indeed more exposure time is needed. In particular more subs. The dithering between images in my opinion is a little small. I'm not sure of the dithering algorithm used in ACP, but I know CCDAutoPilot's does I fine job of dithering. Some subs didn't appear to be dithered at all, hence the noisy data.

Gee, Mike I can't give all my processing tips away.;) OK. Here's what I call "selective contrast masking". Once you're getting close to the end of your processing routing i.e reduced noise etc - Flatten your image in PS. Duplicate this layer (so you have a layer 1 and background). On layer 1, enter curves dialogue (ctrl+m) and make a drastically non-linear curve. It doesn't matter if you white clip the data. I typically leave it clipped, but its a matter of preference/trial and error. Once you made the curve, click ok. Then with layer 1 still selected, blend the data in the multiply mode. From this point you will note you have boosted the contrast of the image, with the background rather dim. Its still contains stars etc. Keep an eye on the histogram from black clipping. Now with this layer, use the colour range tool to select objects from image so you slowly build a selection map. Once the selection map is created and you've verified key features are highlighted, create a selection hide all mask. You may choose to unselect the key foreground object using the lasso tool. This will bring out these features from the background layer. There are a few options with "selective contrast masking", in particular the doing the routing twice with different selection masks. After this, I flattened the image and ran over the individual galaxies using the burn tool and subtle faded the effect. Hope this helps.

I fully concur regarding the 11000 chip. Its not the best when it comes to sensitivity at red wavelengths in particular the all important Ha at 656.3nm. However, it is still the camera of choice for many of the pro's. You can always get around the sensitivity with longer subs/exposure time. I investigated going down the 9000 or 16803 path around 3 months ago for similar reasons (lack of 11000 chip sensitivity), but don't feel I was ready to make the jump. Firstly, you are going to need a decent processing computer to work the output of these cameras. A single image 1x1 bin on the 16803 is ~30Mb. You'll do a lot of thumb twiddling waiting for processing routines to finish. From a single nights worth of imaging, your collected data will range in to the gigabytes worth. I come close to half a gig with the 11000M so capacity is a never ending issue. Secondly, image circle - what's the point of a big chip when you cant evenly illuminate the entire frame. Seems to be a bit of a waste. I wish you the best of luck with your 4" flattener and 155EDF! These chips need an image circle size of no less then 55mm diameter otherwise vignetting will occur. Something the FSQ will easily handle with its 88mm image circle. The bigger the chip, the flatter the field will need to be otherwise it will show. Finally, filters - if you think you are going to use the 50mm circular filters you are kidding yourself. Sure it will work, but at the expense of vignetting. If you are going to do it right, you need a set of square filters. So you'll need to ditch the FLI filter wheel as well as the camera. A set of Astrodon square filters (LHaRGB) will set you back US$2K. Square filters aren't cheap. So ultimately, if you've got a scope that will illuminate the entire chip and prepared to out lay for a new filter wheel and filters - you would have one incredible data collection combo. I look forward to seeing how you get on.

Thanks for the comments Mike.:thumbsup:


Thanks Daniel. :) The background galaxies create an extra dimension.


Thank you all for your comments.

[Bassnut]

Hi Jase

I enjoyed veiwing this image, but given the equipment and expertise, sorry, just not up to the expected Jase standard. Nebulae is somewhat weak and noisy on zooming, even though you got the outer reaches, and stars are slightly wonky, dont know why.

I have the highest regard for your skills Jase, please dont be offended. I could praise everything but wont. Look forward to another try ;-).

Cheers (ive had a couple of drinks, oh dear, what have I done)

[jase]

Thanks for the feedback Fred. :) Appreciate your honesty. I do acknowledge it’s not the best image I’ve put together. It doesn’t do justice for the equipment combination. The data is noisy due to poor dithering and object elevation at time of capture. I pinged Brad an email, but he didn’t respond regarding my dithering concerns. When I blinked the calibrated subs, there were chip defects the showed up in the exact X Y position on each frame. I think what I’m going to do is instead of using the ACP default script for dithering, I will change it to #DITHER 2 or 3. My only concern in doing so is that I loose the guide star between subs. As the scope is guiding on-axis i.e at ~2855mm, the little ST-237 chip only has a 4.4 x 5.8 arcmin FOV. So if I dither too much, losing the guide star is inevitable. Collecting more data is key, but as you are aware this comes at a cost. Actually I don’t even know if I’m enjoying narrow field work so need more of a taste before I make the leap. There is something special about the flat wide fields delivered by the FSQ that draws me back to that instrument.

The disfigured star shapes are from the upscaled RGB data as this was collected at 2x2 not 1x1. I thought I’d capture the chrominance data as binned to improve the S/N ratio, but don’t think it did much anyway. I don’t see any benefits in binning RGB if you have to upscale the data to match the lum. In future I will collect everything as 1x1.

Don't want to share your Helix luminance data you took with the 10" RC with me do you? :P;) kidding...

[Bassnut]

Hi Jase
Use my Helix data?, id be honoured ;-).

Im hoping to avoid the dithering/guidestar loss drama with solid external, id be interested to see your feedback when it comes to fruition.

Im supprised at your RGB binning conslusion, especially with LRGB, this should have nothing to do with distorted stars despite the upscaling, although I guess distortions could be amplified (mind, I have much to learn) . I have found even crappy RGB at 2*2 with a good 1*1 Lum can render excellent results.

Cheers