Border Management Application of Biometrics a Technical Perspective Rick Lazarick, Chief Scientist C

1
Border Management Application of Biometrics a
Technical Perspective Rick Lazarick,
Chief Scientist CSC -
GSS IDENTITY LABS

Panel Biometrics Changing Border Security
6th International Public Safety/Counterterrorism
Conference April 24, 2007
2
Presentation Outline
Scope Border variations ICAO and MRTD
specification for e-Passports Biometric
interoperability Automated face recognition Land
border issues
3
Scope
Question for the Panel What are the critical
elements of border security and can biometric
technologies meet the new standards for
identification and identity management? Focus on
Biometric Technology Emphasis on Usage at the
Border
4
Usage of Biometrics
Identification applies at enrollment and issuance
of passports/visas Relatively infrequent Large
scale search 1-to-n Verification applies at the
border Frequent 1-to-1
5
Border Variations

• Application of Biometric Technology will be
  influenced by
• Land or Port
• Citizen or Visitor
• Voluntary or Mandatory
• Dedicated or Shared
• Careful attention to these qualifiers will
  enhance discussions

6
Border Variations - Examples
CANPASS, NEXUS Both-Citizen-Voluntary-Dedicated
(Iris recognition) US VISIT (initial
rollout) Port-Visitor-Mandatory-Dedicated (Finge
rprints) shared between Canada and US only
7
ICAO and MRTD

• Machine Readable Travel Document
• Specification, basis for e-Passport
• LDS defined to support interoperability (data in
  general)
• Biometric Face (primary) Fingerprint and Iris
  (optional)

8
Passport Reader Interoperability

• Initial emphasis assure interoperability of
  e-Passport chips and readers from multiple
  suppliers
• Essential first step, but does not address
  biometric interoperability

9
Biometric Interoperability

• For Future Shared-Mandatory-Visitor
• (or both)
• This was the intent of ICAO-MRTD?
• Only face recognition
• e-Passport face photo will be available, and
  interoperable
• Standards based photo assurances

10
Automated Face Recognition

• Access to suitable passport photo
• Verification process can be locally defined
  may vary by
• Nation
• Location or location type
• throughput
• lighting

11
Automated Face Recognition Issues
Sensors and acquisition procedures Resolution,
environment controls Matching process Choice of
algorithm (s) Number of samples
processed Fusion potential Note recent
testing results FRVT2006
12
Land Border Biometric Considerations

• Demanding scenario with high throughput, variable
  environment
• Voluntary-Citizen-Dedicated most suitable
• Options for alternate biometrics and integration
  of other technologies
• RFID identification of individuals
• Vehicle identification
• Voice? Self-administered samples?

13
Experience. Results.
Rick Lazarick, Chief Scientist Global Security
Solutions Identity Labs Computer Sciences
Corporation