NGC 2997 - Bunyip Deeeep Field

[David Fitz-Henr]

I finally managed to get all the data needed to process this image in spite of the incessant poor weather :mad2:; it's probably my deepest galaxy image to date and there are quite a few background galaxies in the field.

Note in particular what appears to be a background ring galaxy seen through the outer spiral arms at about 5 o'clock, and also another one at the same angle but 80% further out from the centre of the frame.

Lum-13x12m, R-9x12m, G-9x12m, B-9x12m

Large Image: http://www.pbase.com/image/149199545

[Larryp]

That's superb, David!

[SpaceNoob]

Love the detail and great colour! Wish I could resolve the fine detail you've captured here with my fsq but I think I need more aperture *lol

Is this at f/5? Looks pretty deep and the galaxy is very detailed considering your pixel size. Great image :-)

[strongmanmike]

Great result Dave, certainly has your hallmark detailed look to it ;)

Although and it isn't that big an issue, I see you had a slight registration issue, this has happened to me a couple of times too and makes the stars a little distracting in what would otherwise be a super image. It probably seems like a bit of work but it might be worth reproing to fix that :)

That Bunyip scope performs so well it could almost pass as a fictitious animal in its own right :P

Mike

[David Fitz-Henr]

Thanks Larry!


Thanks Chris; taking the corrector (0.95x) into account the effective focal ratio is f/4.8 (1.22 arcsecs per pixel).


Thanks Mike! Actually, there is no registration issue per se although you may be seeing some differences in diffraction glare due to the fact that I took most of the red on one side of the meridian and the blue on the other side. Due to minor differences around/in the entrance pupil there will always be some variation in diffraction (look at any bright star in an image and you can see that the diffraction pattern is not symmetrical).

[strongmanmike]

Yeh :question: looks like registration error to me :shrug: not in your other images :confuse3: maybe your star saturation technique has just over emphasised the cyan in the glare you mention, hmm?. Anyway, lots of little galaxies in there huh?, great stuff. I noticed that ring galaxy (http://www.pbase.com/strongmanmike2002/image/132326427/original) too a little while ago with the Starfire, very cool :)

Mike

[RickS]

Lovely image, David! The galaxy is fantastic. I did notice the same thing as Mike but it doesn't detract from the main subject.

[alpal]

Hi David,
Excellent image - the best I've ever seen of this dim galaxy -
at about mag 10 it's a tough target.

Yes - the diffraction pattern seems a bit strange but who cares?

You give me motivation to do some more imaging.
Top work - well done.

cheers
Allan

[marc4darkskies]

Beautiful David! Colour processing spot on. Screeds of detail - your trademark now! :thumbsup::thumbsup: I love it!

Nit picking, I do find the magnitude of the "registration" problem a bit distracting but it looks like it would be a royal pain to try and process out since it appears to be inconsistent across the frame :question:.

Cheers, Marcus

[multiweb]

That's an amazing field David. True to the bunyip. Unreal. :thumbsup::thumbsup::thumbsup:

[David Fitz-Henr]

Thanks Mike, Rick, Allan & Marcus!
Actually, on looking more deeply I agree that it was a registration issue. What I said about the diffraction effects are true though, but mainly on the brighter stars (having taken red and blue on opposite sides of the meridian). Looking closely at the other stars I find there is a slight misalignment of the red channel; when I reran Registar and compared I found that the red channel was then properly aligned (must have been a hiccup in Registar the first time :mad2:). Oh well, I may reprocess at a later date.
This may turn out to be fortuitous though, as I have found another minor issue with Registar: The standard interpolation option (Bicubic) resulted in darker pixels in the star images that seem out of place (and would also cause minor colour issues). When I changed to Bilinear, the problem disappears (even though Registar recommends Bicubic)? See the attachment - the star on the left shows some darker pixels when using Bicubic in Registar. Can anyone shed some light on this?


Thanks Marc!

[multiweb]

I find I used to get bright coloured pixels in stars in CCD Stack if I stacked color subs rather than mono per channel. It gets worse when there's not enough pixels per stars. It's the bunyip's fault. Those stars are two damn round and sharp because the light is not spread in too many neibourghing pixels. :P
Might want to try that. I don't know your current flow but try stacking channels separately first then combine colour as the very last step.

[SkyViking]

Fabulours work, I always enjoy these deep galaxy views. Your detail and colouring is impressive, great work as usual. :)
Looking forward to the next one!

[gregbradley]

Mate you are becoming one of my favourite imagers! Sweet! That mirror is a gem - so sharp.

A couple of nitpicks that could help you. There are registration errors in the stars. Your LRGB masters are not properly aligned and are noticeable in the corners. You can see the cyan or red etc sticking out of stars instead of being accurately aligned on top of each other. I use the plug in CCDIS for CCDstack and that stopped any registration issues I was having. The bright blue star middle bottom has an odd cyan colour which is unnatural and is a processing artifact. That colour could be corrected.

The galaxy itself shows fabulous detail and has been nicely sharpened but your sharpening has been allowed to affect the star in that area as well and they are damaged and look oversharpened. When selective sharpening you can click on white or black background tool White reveals and black hides in layer masks. So you select the dropper tool, set it to a small radius and click black on the stars that got oversharpened when you are doing your sharpening mask and before you flatten the layers. This means the stars are unaffected by the sharpening layer and leaves them looking natural. A few little things like that could elevate the image to the next level.

Greg.

[David Fitz-Henr]

Actually I always align/stack mono per channel, and then produce the combined (usually Mean) channel images (ie. L, R, G & B images), and then align these images prior to combining colour, etc. I've sent you some cropped test images for you to test and see if you get the same result/problem in Registar.


Thanks Rolf; my next one may be a repro of this once I get to the bottom of the registration issue!


Thanks Greg. Yes, I've found that the red channel was misaligned slightly for some reason - must have had a hiccup in Registar as I tried Registar again and the result was much better. I also found another minor issue as described above with Bicubic interpolation in Registar creating dark pixels around stars which I am trying to get to the bottom of.
Actually I do protect the stars from sharpening & other filters using layer masks; possibly the misalignment of the red has given this impression of sharpening.

[Stevec35]

Hi David

Registration issues apart it's a fine image of this galaxy.

Cheers

Steve

[clive milne]

David...

Your rendition of 2997 surpasses David Malin's efforts with the AAT.

Well done!

[Shiraz]

an exquisite image of the main galaxy David - top shelf result.
The many small galaxies are very well resolved too.

[Ross G]

Hi David.

That is one beautiful photo!

Sharp, amazing detail and great colours.


Ross.

[David Fitz-Henr]

Thanks Steve.


Thanks Clive, maybe not quite that good but I'll accept the compliment!


Thanks Ray.


... and thanks Ross!