OK, OK, I let the cat out of the bag with this image as I released it on the SBIG forums for a reality check and wake up call. Helps to know if I’m on the right path. A few members here also participate on that forum so I apologise for the duplicate post <yawn>. Anyway, here is one of the reprocessed images in which forms part of the CWAS panel. You’ll have to keep guessing on the other images.
I've added no additional data to this image, I figured 11.5 hours was enough considering the data was of good quality (in this game is it ever enough?). Many may recall the previous image had a disappointing magenta tint. I didn’t bother attempting to restore this, instead I reprocessed the image from scratch. So without further ado, I’m pleased to present the reprocessed version of the
NGC292 - Small Magellanic Cloud
Notes on the reprocessing (concise version);
No recalibration of course, just used the calibrated fits. Registered in Registar and combined in MaximDL using Sigma-Reject. Luminance went through two iterations of deconvolution which was later masked before being introduced to the image providing greater control. I spent most of the time on the colour balance. Using MaximDL colour combine function, I started with the weights 1.5:1.0:1.5 (RGB respectively), then turned on background equalisation. Reviewing the balance, I noticed the Red channel was still too high, so I settled on subtly dropping this, but the blue needed a boost so when with 1.3:1.0:1.6. I made note of these final weights and cancelled the combine operation. I then used pixel math to subtract the background from each channel, but leaving the pedestal value (100 counts). At this point I used pixel math again to add counts to the each individual R, G, B frame. 130%-R, 100%-G, 160%-B. For some reason I doubted by previous g2v balance I had taken a while back. Checking previous note I had made, I soon realised it was pretty close - 1.35:1.0:1.65. Once the individual R, G, B images had the appropriate counts applied to them, I returned to the MaximDL colour combine function and combine the images with a 1:1:1 ratio. This appeared to work well. To emphasis the fainter regions and wisps of nebulosity, I blended 20% of the R channel into the Luminance. Frames matched using the brightness/contrast tools and noise reduction masks applied in PS. Globulars got some minor treatment. Actually, I cropped them and performed a two phase DDP stretch to minimise core burn while showing the full cluster expanse. The two phases were later recombined and layered into the image. Not much else… that was concise eh?
Strangely, I’m not tired of looking at this image, even having spent approximately four to five times the total exposure time processing it! The full size image presented is actually a 70% reduced image. Darn those CWAS print size limitations…a 16x20 300dpi print does not do this justice. Oh well, Enjoy!
Cheers