Iris Recognition Device Usability as Observed in Comparative Testing The International Workshop on Usability and Biometrics 24 June 2008 Michael Thieme.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Iris Recognition Device Usability as Iris Recognition Device Usability as Observed in Comparative TestingObserved in Comparative Testing
The International Workshop on Usability and BiometricsThe International Workshop on Usability and Biometrics24 June 200824 June 2008
Michael ThiemeMichael ThiemeDirector of Special Projects Director of Special Projects
International Biometric GroupInternational Biometric [email protected]
TopicsTopics Iris recognition capture device landscape Application-specific usability Special considerations in mobile collection Evaluating usability through scenario testing Usability-related performance test results Observations on iris usability Potential research areas
About International Biometric GroupAbout International Biometric Group International Biometric Group, LLC
– Founded 1996; 40 employees– Staff in New York, DC Metro, San Francisco– Technology-neutral, vendor-independent
Biometric systems integration– Design, develop, implement biometric identification systems– From mobile data collection to large-scale matching
Biometric R&D and evaluation – 10+ years of biometric R&D and evaluation expertise– Conducted $5m+ in USG-funded biometric R&D
Biometric consulting– SMEs for USG agencies: DHS, Executive Branch, IC– Insight into biometric vendors, industry landscape– Hands-on experience with all biometric technologies
– Tested Panasonic AuthentiCam, LG2200, IrisGuard, JIRIS– Standardized scenario test methodology applicable to fingerprint,
face, voice, signature, hand, vein– First test effort to measure and report false non-match and
failure-to-enroll rates for iris recognition DHS Independent Testing of Iris Recognition Technology
(2004-2005)– Collected approximately 120,000 irises, conducted 2b matches– First test to evaluate interoperability across multiple iris devices– First test to measure statistically significant false match rates
Currently performing operational testing for tactical field deployments of iris recognition devices (GWOT)– Factors: usability, data quality
– Single vs. dual-iris imagers– Simultaneous vs. serial capture– User-adjusted vs. auto-adjusting– Fixed-focal length vs. auto-focus – Volumetric range– Audio and visual prompts– Single-capture enrollment vs. multi-capture enrollment
Arguably greater inter-device variation than in any other modality
Special Considerations in Mobile CollectionSpecial Considerations in Mobile Collection Mobile iris collection a dynamic area
– ~1m irises collected through BAT Adds the operator as a variable
– What prompts are most instructive to an operator?– Does the operator have two hands free?– How close is too close?– Tactile vs. distance-based devices