The SKICAT System for Processing and Analyzing Digital Imaging Sky Surveys
We describe the design and implementation of a software system forproducing, managing, and analyzing catalogs from the digital scans ofthe Second Palomar Observatory Sky Surveys. The system (SKICAT)integrates new and existing packages for performing the full sequence oftasks from raw pixel processing, to object classification, to thematching of multiple, overlapping Schmidt plates and CCD calibrationframes. We describe the relevant details of constructing SKITCAT plate,CCD, matched, and object catalogs. Plate and CCD catalogs are generatedfrom images, while the latter are derived from existing catalogs. A pairof programs complete the majroity of plate and CCD processing in anautomated, pipeline fashion, with the user required to execute a minimalnumber of pre- and post-processing procedures. We apply a modifiedversion of FOCAS for the detection and photometry, and new software formatching catalogs on an object by object basis. SKICAT employs modernmachine learning techniques, such as decision trees, to performautomatic star-galaxy-artifact classification with a > 90% accuracydown to ~1^m above the plate detection limit. The system also provides avariety of tools for interactively querying and analyzing the resultingobject catalogs. (SECTION: Computing and Data Analysis)
|
