IRIS RECOGNITION BORDER-CROSSING SYSTEM IN THE UAE

For the past three years, the United Arab Emirates (UAE) has employed an iris recognition system for all passengers arriving in the UAE by land, sea or air. By looking at an iris camera for a second or two while passing through immigration control, each passenger's iris patterns are encoded mathematically and the resulting image is sent over a distributed communications network to a central database where they are compared with the images from 420,000 people who were expelled from the UAE for various violations. Less than one second is required for each image to be compared against the full database. More than 9,500 people have been caught traveling with forged identities by this system. This article describes the system in detail, including the critical features of the system architecture and the algorithm for encoding and comparing iris patterns. The system operates in an "all against all" identification search mode. The decision thresholds are automatically adjusted according to the size of the enrolled database and thus the number of comparisons performed, so that there is no net accumulation of probability of false matches. This makes the system extremely efficient and accurate. All of the matches made so far between arriving passengers and violators in the database have been confirmed by other records.

• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/oclc/37505394
• Corporate Authors:

  Russell Publishing Limited

  Court Lodge
  Hogtrough Hill
  Brasted, Kent  United Kingdom  TN16 1NU
• Authors:
  • Daugman, J
• Publication Date: 2004

Language

• English

Media Info

• Features: Figures;
• Pagination: p. 49-53
• Serial:
  • International Airport Review
  • Volume: 8
  • Issue Number: 2
  • Publisher: Russell Publishing Limited
  • ISSN: 1366-6339
  • Serial URL: http://www.russellpublishing.com/pages/airport/airport1.html

Subject/Index Terms

• TRT Terms: Algorithms; Biometrics; Databases; International borders; Security; System architecture
• Candidate Terms: Iris recognition
• Geographic Terms: United Arab Emirates
• Subject Areas: Aviation; Data and Information Technology; Security and Emergencies;

Filing Info

• Accession Number: 00979224
• Record Type: Publication
• Files: TRIS
• Created Date: Sep 12 2004 12:00AM