PixInsight > Gallery
NGC 3810 first cut
dayers:
All,
I have been remotely imaging NGC 3810 for the past month or so using the University of Arizona SkyCenter 32 inch Schulman. Exposure: 4.3 hours Luminance and 2.7 hours total RGB. The raw files were calibrated in CCDStack. I moved to PI for StarAlignment, ImageIntegration, and subsequent processing of the LRGB you see here.
http://orlop.net/images/NGC3810_LRGB_prelim20120427/
Before I try for an improved image, I would love to hear about areas that look like they could use a little more work (or a lot more, for that matter!) In particular I am concerned about the color fringing around the brighter stars and about the background (about k:0.08). I applied some mild background de-noising using MMT. As for the galaxy itself, I boosted the saturation with LRGBCombination, gave it a sniff of HDRMT, and at the very end touched it up with a mild S-curve in CurvesTransformation. And I used CloneStamp on a couple of blotches, but I won't tell you where. I tried to keep it plain and simple.
So there you are. Please have at it. I'm still scrambling up the PI learning curve, and thanks.
Dave
Nocturnal:
I like it a lot Dave! The number of small background galaxies is amazing!
sreilly:
Dave,
Wonderful image but I'm not sure about the alignment. Looks like green/blue on the top of the stars and red at the bottom. Either the alignment is off or the stars are bloated quite a bit. I'm thinking star alignment is the issue. Just for kicks, are all the images binned 1x1 or is the color 2x2 and luminance 1x1? I've taken to imaging using all 1x1 images but that's a choice I made. Either way, make sure you are aligning all your images to a single reference image. Being all the 2x2, if that's the case, aligned to say the 1st red image. That means all color, all red, green, and blue images are aligned in a single run to the 1st red image of the stack. Save these and go from there. If all the images are binned 1x1 then include the luminance images with all the color images. Hope I'm clear on this. I use to do all the red, then green and blue and then align the master images but it takes longer.
For my taste the background is way too dark and would probably reveal way more detail if lightened up using the histogram tool. But that's my taste.
Steve
dayers:
Thanks Sander and Steve,
The color was binned 2x2 and the luminance 1x1. The whole batch of calibrated files, color and luminance, were aligned together using the lowest FWHM luminance as the reference. Next time around I'll try aligning color separate from luminence using the "best" color image as reference for all the RGB. Somehow I think you may be right about alignment being the source of the color fringing. Thanks for your comments.
As for background level, I thought that darker background would allow the most "down deep" detail to appear. No? I'll try background of around 0.1000 to 0.1200 and compare. That's about where I started out, but in a last minute CT S-curve I lowered the background a little and kept the galaxy core brightness the same to nudge the contrast up a little.
Thanks again for your observations. I really appreciate it.
Dave
ajbarr:
Dave I think you should try HDR Multiscale transform which will increase detail in the core significantly. I put a luminance mask on it and ran HDR and it seems to help quite a bit, at least in my opinion.
Navigation
[0] Message Index
[#] Next page
Go to full version