Nist Evaluating Iris Recognition

Iris recognition is a potentially valuable biometric (a characteristic such as fingerprints that can be used to identify a person). Previous evaluations of iris recognition have concentrated on the performance of systems. The ICE is the first evaluation designed to measure the accuracy of the underlying technology that makes iris recognition possible. The goals of the ICE are to (1) promote the development and advancement of iris recognition and (2) assess the technology’s current level of performance. This will be accomplished in two phases. Phase I—which will be conducted from now until the end of the year—will seek researchers and developers from industry, research institutions and academia who interested in participating in iris recognition “challenge problems.” Taking part in these “challenge problems” (designed to promote technology development) will give participants the opportunity to improve their current performance rates and help prepare them for ICE Phase II. Phase II—tentatively scheduled for the first quarter of 2006—will give iris recognition system developers an opportunity to take part in a large-scale, independent evaluation. To guarantee accurate assessments, the ICE will measure iris matching performance with sequestered data (iris images not previously seen by the participants). A standard dataset and test methodology will be employed so that all participants are evaluated evenly.