Seeking advice on data rejection for integration

[joelshort]

I'm seeking advice on data rejection while stacking.  I have some images that have a pronounced reflection from a nearby star.  The reflection looks like a very defocused star and it moves relative to the movement of the bright star on the CCD plane.  In other words, when I dither the exposures the reflection moves slightly relative to the star.  I'm trying to figure out a way to reject the reflections during image integration but have not had much success, no doubt because I'm new and have no idea what i'm doing  .

Thanks for any advice. 

[pfile]

i think this could be difficult as the reflection is not super bright (the way a hot pixel would look) and the reflections probably overlap significantly from frame to frame, right? meaning the intersection of all the reflections is going to look more or less like 'real' data to ImageIntegration.

rob

[joelshort]

Uhh...yeah.

The reflections do look enough like real data to cause problems.  But they actually don't overlap very much.   I don't really understand the different rejection methods (new to PI) and when I've tried to integrate it will eliminate about 80% of the reflections.  The real problem of course is trying to eliminate the reflections in the first place, but that's another topic.

[pfile]

II is one of the tools that has documentation. it's rather technical but if you click on the little dog eared page at the bottom of the ImageIntegration pane, you'll see the documenation.

basically the rejection methods differ mainly in how they compute what is an outlier pixel, and what they do to replace discarded data. probably the thing to do here is to keep pushing the sigma high slider down and re-integrate, paying attention to the rejection maps to make sure you are only going so far as to reject the reflections.

rob