The OPTID Database: Deep Optical Identifications to the IRAS Faint Source Survey

We use a new, robust, method to estimate the identification probabilities of optical matches from digitized plate catalogs (COSMOS/UKST Catalog of the Southern Sky Version 2, Yentis et al. 1992; The Guide Star Catalog Version 1.1, Lasker et al. 1990; The Tycho Input Catalog, Egret et al. 1992; The APM Northern Sky Catalogue, Irwin, Maddox and McMahon 1994) to sources in the IRAS Faint Source Survey (FSS; Moshir et al. 1992), including both the Catalog (FSC) and the Reject File (FSR), utilizing a new random matching procedure with the advantages that it: (1) eliminates systematic uncertainties due to many problems, such as uncertainty in N(m); variations across the optical plate of magnitudes, plate limits or N(m); misclassification of stars and galaxies; the assumption of Gaussian error ellipses, etc. and (2) properly calibrates the identification probabilities. We find that at high SNRs and high galactic latitudes essentially all IR star-colored sources have an optical identification with P id > 99%. At high SNRs and high galactic latitudes, ~ 90% of all IR galaxy-colored sources have an optical identification with P id > 90%.