View Single Post
  #1  
Old 27-11-2008, 07:12 PM
jase (Jason)
Registered User

jase is offline
 
Join Date: Sep 2006
Location: Melbourne, Victoria
Posts: 3,916
Gravitational Galaxies in Grus

Hi All,
I’m pleased to present “Gravitational Galaxies in Grus”, the latest collaborative effort between Alvin of Lightbuckets and myself. This area of the night sky has intrigued me for sometime due to the interesting characteristics of the interacting triplet.

About the target;
Located in the constellation Grus, three spiral galaxies located very close to each other interact. When galaxies are close enough, a mutual gravitational attraction can result, often altering the appearance and in some cases the composition of their systems. An example of this is the galaxy center of frame, NGC7582. This barred spiral galaxy exhibits tidal tails extending towards the two neighbouring galaxies (NGC7590 and NGC7599) at top left and another galaxy below which is not present in the frame (NGC7552).

About the image;
The image is a straight LRGB composite with a total exposure time of 12.6 hrs (L:260min,R:160min,G:150min:B:190mi n). Data acquisition performed by Alvin on the Lightbuckets 14.5” RC (3315mm F/9) in Pingelly, WA. He came up with a unique composition of the trio which has progressively grown on me, with the center showpiece being NGC7582. The other interacting galaxies (NGC7590 and NGC7599) add a nice dimension at top right of frame. The “void” to the lower right diverts ones attention and allows the numerous faint background galaxies to really come through, offers an interesting perspective. As the RC FOV is small, the forth member to complete the quartet, NGC7552 is off the frame. A mosaic at this FL would have been magical. I took a basic approach to processing this image with the goal of maximising the background galaxies and displaying a natural colour saturation of the foreground objects. After the usual calibration, registration and creation of masters, I duplicated the luminance data and pushed it through to different iterations of deconvolution. The first was relatively light, while the second luminance was strong. The latter revealed some incredible features on the oversampled data, but destroyed stellar profiles. The lightly deconvoluted image simply provided tighter stars and was used as the main luminance. This was heavily stretched in DDP before being imported into PS to bring out the faint background galaxies. The quality data stretch quite well, with very little noise exhibited. The stronger deconvolution was simply imported straight into PS and introduced using a mask at 30% opacity. I blurred the mask to provide a smooth transition. This provided control of the integration of the rich details for the three foreground objects. The two layers were flattened with a very light USM applied. Similarly, two copies of the RGB combined master were created. One imported directly into PS and was stretch manually using levels and curves. This was also adjusted to get the colour foundation correct. The luminance was then introduced at 60% opacity. The second RGB master had its saturation boosted to ~190% and pushed through a hard DDP stretch, almost to the point where highlight clipping was experienced. This master was imported into PS and introduced as a softlight blend at approximately 70%. Luminance was then increased to 85%. Minor colour balance tweaks applied to manage the softlight blend impact to the layers. The softlight blend layer had the minimum filter applied with the fill effect dropped back to 25%. Yes, you heard…I used the ghastly minimum filter, but when used appropriately, it can work wonders. This gave a richer appearance to star colours. A minor gradient from the green channel was experienced along the lower parts of the frame, but cleaned up with GradientXterminator. The image was then flattened and black point set appropriately. While its tempting to make the background ultra dark to maximise the faint background galaxies, this resulted in too much contrast for this image, leading to an unnatural transition between features. Minor noise reduction invoked via invert mask. Could have perhaps gone harder.

I’d like to thank Alvin for the opportunity to process this data set. The area has been on my seasonal target list for sometime, but never intended to nail it at 3315mm FL. The image scale delivered by the FL provides some “punchy” details on the triplet.

Couldn’t think of a better way to mark the 2,000th post than with a pleasing image to share amongst like-minded souls. Anyway, enough yabbering. Thanks for having a look - I hope you enjoy as much as I did processing it! All comments welcome.
Reply With Quote