Quote:
Originally Posted by Placidus
Colin, that worked really well.
I spend about an hour a day thinking wishfully but unsuccessfully about the underlying maths for removing stars. Don't want to use someone's secret proprietary technique, no matter how well it works, I want to be able to write the code.
I wonder if part of my problem is that we do very long subs where the stars burn out, so that all information under the star is lost. A bit like the difference between trying to work out what's under a piece of grey transparent plastic, versus what's under a steel disc.
Any thoughts?
Best,
Mike
|
I have been thinking about it, there is no viable information under any stars. When you have a nebulae ADU count of 1000 even fun stars are likely to be closer to 5-15,000 ADU so very much dwarfed.
Mathematically, I think what you need to do is statistical background broadening [made that up] in a local area (64 pixel square?). You take the statistical average of a lower limit in an area, takes the average background and excludes stars which are generally significantly more intense. You take this value and then average every unmasked pixel by that value. The background (nebulosity) is largely unchanged BUT stars are hit hard, the brighter the harder.
Quote:
Originally Posted by multiweb
Looks excellent Colin. Well done. 
|
Quote:
Originally Posted by atalas
Very cool Colin....I could never get the hang of doing that!well done.
|
Thanks

I really needed to clone out some of the brighter nebulosity areas out of the star mask but as I need to retake the data anyway... I got lazy