ASPCS
 
Back to Volume
Paper: Bayesian Belief Networks for Astronomical Object Recognition and Classification in CTI-II
Volume: 376, Astronomical Data Analysis Software and Systems XVI
Page: 413
Authors: Ritthaler, M.; Luger, G.; Young, R.; McGraw, J.; Zimmer, P.
Abstract: The University of New Mexico (UNM) is currently designing and building the CCD Transit Instrument II (CTI-II: McGraw et al. 2006), a 1.8-m transit survey telescope. The stationary CTI-II uses the time delay and integrate readout mode with a mosaic of CCDs to generate over 100 gigapixels per night which are required to be analyzed within a day of observation. We are attempting to develop robust machine learning techniques that use multiple scientific and engineering data streams to classify objects within an image frame and the image frame itself. We propose the use of Bayesian belief nets as both classifiers and as tools to integrate and explore the data streams. This initial report explores the use of Bayesian networks as source/noise separators.
Back to Volume