Using the Sub Frame Selector Script with multiple groups

[Batch]

I am hoping that someone suggest the best way to combine separate groups of calibrated subs that have been separately graded in the Sub Frame Selector Script.

I use the weighting calculated by the script when integrating the images (Weights - FITS Keyword) so if I am combining two imaging sessions can I load both sets of separately calibrated subs into the Imaging module and use the FITS keyword weighting parameter? Or is it better to run all of the calibrated subs from both sessions through the script a second time and then integrate them? As well, as the individual subs from each session are different in terms of time and iso will this effect the grading by the script if they are grouped together? Thanks

Bill

[dnault42]

This depends on how you graded each group.  What weighting and rejection expression did you use?  Were they different for the groups?  Setting the appropriate gain and image scale should help to equalize the measured data so that the comparison is valid.  You'd need to know the gain at each of the different ISO settings.  It's easy to calculate if you have a couple flats at the different ISO levels by using the FlatSNREstimator script by Mike Schuster.

Regards,
David

[mschuster]

I agree with David, it depends.

I recommend running the script a second time with all the frames and check to see if any of the frame weights changed. If the weights did not change, then of course there was no need to run the script a second time. The result of this test will depend on the expression.

Weighting, normalizing and integrating different ISO's and times requires some care.

Mike

[Batch]

Thank you very for the feedback.

 I have been using (2 * SNRWeight) - FWHM as my weight expression and I have not been changing the camera gain when running the script. I'll run the script twice and see if there is a change in the weights. Thanks again.

Bill

[mschuster]

Hi Bill,

Your expression should not change, as it depends only on individual frame statistics, and not frame set statistics.

You might want to try SNRWeight / FWHM^2. It avoids having to find mixing constants which are somewhat frame set variable, and also has a more sound theoretical basis IMO.

Any expression using SNRWeight and FWHM should work well across sets with varying ISO and time. This is a useful property.

Thanks,
Mike

[Batch]

Thanks Mike, I really appreciate you taking the time to give me a hand. I'll change the expression and give it another go.

Bill