I'm trying to follow Adam Block's technique outlined here
https://pixinsight.com/forum/index.php?topic=11609.0to do selective rejection to exclude pixels from a dust donut that showed up in my images from Night 1 but not Night 2. Since the exposures were taken on different nights, they have a different background brightness and different gradients. In the area around the dust donut, Night 1 has a higher brightness. By using Adam's technique, I can exclude the dust donut area of Night 1 from the integration, but if the rest of Night 1 goes into the integrated image, the area around the dust donut will be brighter than the dust donut area itself, because Night 1 was just brighter in that whole area in general than Night 2.
I thought Local Normalization might solve this: before applying Adam's technique to paint out the dust donut, register both nights and make LNorm files. Then in the integration process, use the Night 1 images that had the dust donut painted out, but use the LNorm files that were created from the registered images that did NOT have the dust donut painted out. Integrate Nights 1 and 2 with the appropriate LNorm files, and you should get an integration that blends smoothly into the dust donut area.
The thing is that ImInt won't let me apply LNorm files not created from precisely the same images as used in the integration. Is there a way to fool ImInt into allowing this?
Kevin
