More Cocoon Nebula experimentation
The stars in your images do not look bloated to me, comparing them to other high-end imagers on the web. They look relatively large due to the image scale that you are working at. Stars that are bloated by differing exposures from the bright foreground stars verses faint background nebulous detail can be shrunk early in processing using: 1) Clip-stretching with a maximum threshold reduction 2) Gamma brightness stretching [then stack] 3) Richardson-Lucy deconvolution 4) Combinations of 2 and 3. The main trick appears to be only clipping and stretching the image to reduce only the maximum side of the brightness histogram. That reduces the number of pixels displaying for the bright stars but does not reduce the detail in the faint nebula. But, not being an experienced imager, I'm hesitant to discuss it. See notes at - http://gallery.utahastronomy.com/v/kurt/KafAIP4WIN/BloatStar/ - Kurt _______________________________________________ Sent via CSolutions - http://www.csolutions.net
On 12 Oct 2007, at 23:01, Kurt Fisher wrote:
The stars in your images do not look bloated to me, comparing them to other high-end imagers on the web. They look relatively large due to the image scale that you are working at.
That makes sense. A couple of others have made the same observation so I had another look at Tyler's DSO pictures (The Gold Standard on this list as far as I'm concerned). Turns out his pictures with the tiny stars were shot with scopes of less than 1,000 mm focal lengths. Mine were shot at nearly 4,000. So I've mounted a C-5 piggyback on the C-14 and will try with that. The C-5 is native 1,250 mm and I can cut that further with various focal reducers. If I get anything with that set up that's worth showing I'll post them for comparison.
Stars that are bloated by differing exposures from the bright foreground stars verses faint background nebulous detail can be shrunk early in processing using:
1) Clip-stretching with a maximum threshold reduction 2) Gamma brightness stretching [then stack] 3) Richardson-Lucy deconvolution 4) Combinations of 2 and 3.
The main trick appears to be only clipping and stretching the image to reduce only the maximum side of the brightness histogram. That reduces the number of pixels displaying for the bright stars but does not reduce the detail in the faint nebula.
Thanks for the advice but I've no idea what any of that means. :) Still, I'll keep plugging away. Cheers, patrick
participants (2)
-
Kurt Fisher -
Patrick Wiggins