You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Megan rates the Voronoi tessellation source finder, so will probably use that. Ultimately I want DAXA to be able to run source detection for all the telescopes, but the actual source detection would ideally live in another module which can also (to some degree) make fake observations in order to be able to easily build selection functions.
The text was updated successfully, but these errors were encountered:
First, must decide which techniques to focus on - could well end up implementing all three(?) that they offer in CIAO for now. Whatever we do though will likely not be forever, as eventually another package will be created dedicated to source finding and selection function generation.
I've heard good things about the Voronoi tessellation CIAO tool, and it is fairly computationally efficient, so might well do that one.
Had a go with the voronoi tessellation - on ObsID 3185 (of my favourite A907). Ran pretty fast, though it crashed the first time I ran it because of a stack size limit on the shell I think? (ran ulimit and it worked the second time).
Does a decent job - certainly detects A907 and many of the outlying point sources:
However it also gets a tonne of spurious sources around the edge of the A907 detection - and misses a couple of point sources that I know are within the A907 extended emission (both from my knowledge of this cluster, and you can see them).
Possibly could fiddle with the settings to improve it, but maybe this isn't the best method for wide application? I wish XAPA were generalised, but that won't be ready for an age.
Megan rates the Voronoi tessellation source finder, so will probably use that. Ultimately I want DAXA to be able to run source detection for all the telescopes, but the actual source detection would ideally live in another module which can also (to some degree) make fake observations in order to be able to easily build selection functions.
The text was updated successfully, but these errors were encountered: