[News] USM ongoing activities and AI's

[Roeland Rengelink]

Dr. Mark Neeser wrote:

>Ongoing Astro-wise Activities at USM:
>
>Current WP2 activities
>----------------------
>
>- creation of stand-alone C program for cosmic-ray detection
>  and correction on individual frames.  Program uses multi-variate
>  fitting of Gaussian profiles.  Although very CPU intensive this
>  method has been found to be very reliable, even in the case of
>  strongly extended cosmic-ray features.  Current test show that
>  with Omega-cam like parameter settings, one can expect a performance
>  of about 15 seconds per CCD on a single node.
>  
>
This is actually comparable to the time that it takes to run sextractor 
with a comsic retina filter to produce a cosmic ray image. We therefore 
may be interested to use this program as an alternative in standard 
pipeline operations. Especially since we would really like to to 
coaddition using the (linear) weighted average, and, hence, cannot use 
lmedian averageing to get rid of cosmics in coaddition.

Very good news indeed.

Roeland