This is a confirmed bug in ImageIntegration, which has been detected by Harry Page (thanks Harry!):
- Pixel rejection does not work correctly for the reference image (the first one in the list of integrated images) when there are one or more images with negative zero offsets. In general, pixel rejection is excessive for the reference image under these conditions.
Workaround (unique): Select a reference image that leads to positive zero offsets for the rest of integrated images. This is very easy to find: just select as reference the image that has a larger median value (use Statistics).
I have managed to identify and fix this bug today. It has been fixed in PixInsight 1.5.6, which will be released during tonight and tomorrow.
I am sorry for the problems this might have caused to you.
