Biometric Technology for DLID An introduction to the science (as applied to Canadian requirements) Ian Williams Principal www.idsysgroup.com Canada Day at DLID Summit, Houston, Texas, February 29 Canada Day at DLID Summit, Houston, Texas, February 29 th th , 2004 , 2004
30
Embed
Biometrics for DLID - Identity Systems Group - DMV DL/ID Experts
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
Biometric Technology for DLID
An introduction to the science(as applied to Canadian requirements)
Ian Williams Principal
www.idsysgroup.com
Canada Day at DLID Summit, Houston, Texas, February 29Canada Day at DLID Summit, Houston, Texas, February 29thth, 2004, 2004
• 2
Biometric Technology for DLIDBiometric Technology for DLID
Biometrics: DefinedBiometrics: Defined
Automatically recognizing a person using Automatically recognizing a person using distinguishing traits. (A narrow definition)distinguishing traits. (A narrow definition)
Biometric technologies are defined as automated methods of identBiometric technologies are defined as automated methods of identifying or verifying the ifying or verifying the identity of a living person based on physiological or behavioralidentity of a living person based on physiological or behavioral characteristics.characteristics.
Recently, this was modified to include chemical attributes (DNA)Recently, this was modified to include chemical attributes (DNA) Nov. 2003Nov. 2003
• 3
Biometric Technology for DLIDBiometric Technology for DLID
Biometric Market TodayBiometric Market Today
What are theyWhat are theyandand
How do you choose?How do you choose?
16+ Technologies16+ Technologies
540+ Biometric Vendors540+ Biometric Vendors
8 Technologies8 Technologies
20+ Biometric Vendors20+ Biometric Vendors
What are theyWhat are theyandand
How do you choose?How do you choose?
1997 – Snap Shot
• 4
Biometric Technology for DLIDBiometric Technology for DLID
Biometric TechnologiesBiometric Technologies
••Iris Recognition Iris Recognition
••Finger ScansFinger Scans
••Hand Geometry Hand Geometry
••Facial RecognitionFacial Recognition
••Signature DynamicsSignature Dynamics
••Voice DynamicsVoice Dynamics
••Retinal ScanRetinal Scan
••Keystroke DynamicsKeystroke Dynamics
Physical biometrics apply to DL/IDPhysical biometrics apply to DL/IDBehavioral is difficult to enroll for DL/IDBehavioral is difficult to enroll for DL/ID
• 5
Biometric Technology for DLIDBiometric Technology for DLID
Biometric TerminologyBiometric Terminology
Verification Verification 1:11:1Measurement of an identity against a Measurement of an identity against a single claimed identity single claimed identity
IdentificationIdentification 1:N (many) 1:N (many) Measurement of a single identity Measurement of a single identity against multiple identitiesagainst multiple identities
Requires identification, to ensure that I am enrolled only once.Requires identification, to ensure that I am enrolled only once.
• 6
Biometric Technology for DLIDBiometric Technology for DLID
DL/ID Biometric UsesDL/ID Biometric UsesVerificationVerification –– Match the user biometric to a Match the user biometric to a single claimed identity. (renewal / resingle claimed identity. (renewal / re--issuance)issuance)
DL/ID requires biometric that can perform BOTH verification & idDL/ID requires biometric that can perform BOTH verification & identificationentification
IdentificationIdentification –– Compare the user biometric to Compare the user biometric to all others in database to ensure they are not all others in database to ensure they are not previously enrolled. (enrollment)previously enrolled. (enrollment)
DL/ID DatabaseDL/ID Database
Am I Am I -- this person?this person?
Am I Am I -- ANY of these people?ANY of these people?
• 7
Biometric Technology for DLIDBiometric Technology for DLID
Iris RecognitionIris Recognition
Video Capture Trabecular MeshworkVideo Capture Trabecular Meshwork Optical “Fingerprint”Optical “Fingerprint” Iris Code Record Iris Code Record
How it worksHow it works
Issues/ObstaclesIssues/Obstacles•• Expensive camerasExpensive cameras•• Sole source worldwide licencing Sole source worldwide licencing •• Enrolment limitationsEnrolment limitations•• Distance for capture (< 4 feet)Distance for capture (< 4 feet)
• 8
Biometric Technology for DLIDBiometric Technology for DLID
Iris RecognitionIris Recognition
Iris recognition is suitable for bothIris recognition is suitable for both IdentificationIdentification & & VerificationVerification
Features/BenefitsFeatures/Benefits•• Highest accuracy (1:N)Highest accuracy (1:N)•• Constant throughout adulthoodConstant throughout adulthood•• No physical damage (picture not scan)No physical damage (picture not scan)•• Requires cooperation Requires cooperation
Iris sampling offers more reference coordinates than any other biometric. Mathematically, it therefore has a higher accuracy potential than any other biometric.
Iris recognition has never had a FALSE match
• 9
Biometric Technology for DLIDBiometric Technology for DLID
KPN Telecom (The Netherlands)KPN Telecom (The Netherlands)U.S. House of Representatives, Legislative U.S. House of Representatives, Legislative
Counsel (DC) Counsel (DC)
United Arab Emirates United Arab Emirates --every day about 2 Billion every day about 2 Billion
iris comparisons are iris comparisons are performedperformed
• 10
Biometric Technology for DLIDBiometric Technology for DLID
ClientClient: Canada Customs and Revenue (CCRA): Canada Customs and Revenue (CCRA)
ApplicationApplication: Border Control : Border Control –– CANCAN--PASS. PASS. Eight airports starting with Toronto and Eight airports starting with Toronto and Vancouver in Q4 ’02. Trials with U.S. at Ottawa Vancouver in Q4 ’02. Trials with U.S. at Ottawa and Montreal in 2003.and Montreal in 2003.
System IntegratorSystem Integrator: IBM Global Services: IBM Global Services
Background: “Background: “Iris is your passport.” CCRA will Iris is your passport.” CCRA will charge $50 annual fee. Card based.charge $50 annual fee. Card based.
Number of usersNumber of users: 100,000 : 100,000 –– largest public largest public sector use of Iris thus far in Canada.sector use of Iris thus far in Canada.
Canada Customs & RevenueCanada Customs & Revenue
• 11
Biometric Technology for DLIDBiometric Technology for DLID
Finger Scanning is suitable for bothFinger Scanning is suitable for both IdentificationIdentification & & VerificationVerification
* Not all finger scanners suitable for 1:N identification, due t* Not all finger scanners suitable for 1:N identification, due to low resolutiono low resolution
• 13
Biometric Technology for DLIDBiometric Technology for DLID
Finger Scan ApplicationsFinger Scan Applications
DL/ID Cards
Arkansas – V
California
Colorado
Georgia
Hawaii
Texas
W. Virginia
Social ServicesTexas
ConnecticutNew YorkToronto*
Secure AccessMicrosoftOracle
New York Stock ExchangeNew York Police Department
Financial Services
Banking ATM
Internet Banking
Retail Transactions
Smart CardSmart Card
Finger ScanFinger Scan
*System withdrawn*System withdrawn
• 14
Biometric Technology for DLIDBiometric Technology for DLID
Biometric Technology for DLIDBiometric Technology for DLID
Hand GeometryHand Geometry
Features/BenefitsFeatures/Benefits•• ReliableReliable•• NonNon--intrusiveintrusive•• Small record size (9 bits)Small record size (9 bits)•• Successful deploymentsSuccessful deployments
Hand Geometry is suitable for Hand Geometry is suitable for Verification Verification only.only.
Hand Geometry is NOT suitable for Hand Geometry is NOT suitable for IdentificationIdentification..
• 16
Biometric Technology for DLIDBiometric Technology for DLID
Hand Geometry ApplicationsHand Geometry Applications
Identification CardsIdentification Cards
INSPassINSPass
Time & AttendanceTime & Attendance
CocaCoca--ColaCola
First Bank of MinnesotaFirst Bank of Minnesota
Secure AccessSecure AccessIBMIBM
ExodusExodus
Sandia National LabsSandia National Labs
San Jose State UniversitySan Jose State University
• 17
Biometric Technology for DLIDBiometric Technology for DLID
Facial recognition is suitable for Facial recognition is suitable for VerificationVerification..
Facial recognition is a useful tool for assisting Facial recognition is a useful tool for assisting IdentificationIdentification..
* Images in databases may not be suitable for automatic template* Images in databases may not be suitable for automatic template extraction.extraction.
• 19
Biometric Technology for DLIDBiometric Technology for DLID
Facial Recognition AppsFacial Recognition Apps
Identification CardsIdentification Cards
West Virginia Driver LicenseWest Virginia Driver License
Illinois Driver LicenseIllinois Driver License
Alberta Operators LicenceAlberta Operators Licence
•• Estimated 2 billion images in ID databases around the world. Estimated 2 billion images in ID databases around the world.
•• Innocuous in a DL/ID environment. Innocuous in a DL/ID environment.
NegativeNegative•• FRVT 2002 indicates questionable accuracy claims.FRVT 2002 indicates questionable accuracy claims.•• Disguises can easily defeat facial recognition.Disguises can easily defeat facial recognition.•• Lighting, age, glasses, and head/face coverings Lighting, age, glasses, and head/face coverings –– impact.impact.•• Aging of comparison images drastically reduces match probabilityAging of comparison images drastically reduces match probability..
Regarding FR Accuracy:
…this seems to throw into question some of the recently announced projects for use of facial recognition to prevent multiple enrollments in large databases.
Dr. Jim Wayman - San Jose State University Biometrics in Human Services Volume 5, Issue 3
• 21
Biometric Technology for DLIDBiometric Technology for DLID
Facial Recognition ContinuedFacial Recognition Continued
Facial Recognition (FR) is the comparison of a face to one or more reference faces, measuring for similarity.
The human brain, conditioned from birth to recognise faces, The human brain, conditioned from birth to recognise faces, compensates for appearance differences caused by weight gain/loscompensates for appearance differences caused by weight gain/loss, s, hairstyle, aging, facial hair, makeup or other subtle physical nhairstyle, aging, facial hair, makeup or other subtle physical nuances, uances,
more effectively than any current machine technology. more effectively than any current machine technology.
Facial Recognition (FR), like filtering, is a useful tool to narrow searches in a large database of images. Given the renewal cycles and magnitude of DL/ID image databases, FR alone
would be insufficient as the biometric for One-Driver-One-Record (1D1R) purposes.
• 22
Biometric Technology for DLIDBiometric Technology for DLID
Canadian Biometric InitiativesCanadian Passport – 2004
Facial recognition - (+ secondary biometric?) Smart chip to be added to passport and to include biometric templates (ICAO standard).
Toronto Welfare 1997-1999Finger scanning – Designed to prevent multiple enrollment of welfare recipients. Program abandoned in 1999.
CIC – National ID card 2003 - 2010Undecided – current House of Commons debate whether to establish a new National Identity Biometric Card.
Alberta Operator Licence - 2003Facial recognition – Initially used as a verification check, may attempt to utilize as an identification application.
CANpass - 2003Iris recognition – Used to allow rapid cross border travel. Uses an optical card to store Iris template.
CATSA - 2004Finger/Iris - New program to be deployed 2004 to secure airport areas using biometric smart card and gate access for employees.
• 23
Biometric Technology for DLIDBiometric Technology for DLID
Choosing a TechnologyChoosing a Technology
IdentificationIdentification
VerificationVerification
EffortEffort
IntrusivenessIntrusiveness
CostCost
Acceptable Zone
FIVE FACTORSFIVE FACTORS
• 24
Biometric Technology for DLIDBiometric Technology for DLID
Data ConsiderationsData Considerations
Match your biometric technology needs with machine read needs, aMatch your biometric technology needs with machine read needs, and nd consider the costs.consider the costs.
Unlike a PIN, a user Biometric cannot be replaced. Therefore template/data stored on cards MUST be secure from theft or wrongful access.
Biometrics stored in Machine Readable Documents (MRD) will likelBiometrics stored in Machine Readable Documents (MRD) will likely require y require mandatory encryption to meet Canadian privacy requirements.mandatory encryption to meet Canadian privacy requirements.
• 26
Biometric Technology for DLIDBiometric Technology for DLID
Biometric Industry FactBiometric Industry Fact
•• Iris is the number one technology in accuracyIris is the number one technology in accuracy
•• Finger is biometric leaderFinger is biometric leader (market share)(market share)
•• Facial is an “easy” addition to DL/IDFacial is an “easy” addition to DL/ID
•• Compounded biometrics for accuracy (US VISIT; UK ID)Compounded biometrics for accuracy (US VISIT; UK ID)
vvStore on card? Store on card? –– “Opt“Opt--in” for public approvalin” for public approval
vvAdd / Improve filtering and dB search methods to improve identifAdd / Improve filtering and dB search methods to improve identif ication search when using ication search when using no biometric or less accurate technologies no biometric or less accurate technologies
vvUse 4 key data search on IRE (with biometric option)Use 4 key data search on IRE (with biometric option)
• 29
Biometric Technology for DLIDBiometric Technology for DLID
Biometrics: RealityBiometrics: Reality
Facial is a useful tool to improve the identification search proFacial is a useful tool to improve the identification search process.cess.
Biometrics Biometrics cancan ensure the verification of an individual to their own record.ensure the verification of an individual to their own record.
Iris recognition is likely the only biometric technology that caIris recognition is likely the only biometric technology that can singularly meet the n singularly meet the 1:N need for identification in North America DL.1:N need for identification in North America DL.
Reality CheckReality Check
vvMost biometric technologies can be dupedMost biometric technologies can be duped
vvNo single biometric can enroll ALL applicantsNo single biometric can enroll ALL applicants
vvNo biometric can prove 100% accuracyNo biometric can prove 100% accuracy
vvBiometrics will not GUARANTEE multiple enrollee preventionBiometrics will not GUARANTEE multiple enrollee prevention
vvHuman intervention WILL be required to resolve probable matchesHuman intervention WILL be required to resolve probable matches
! !
• 30
Biometric Technology for DLIDBiometric Technology for DLID
Get more info here:Get more info here:
www.idsysgroup.comwww.idsysgroup.comBiometric Technology Resource CentreBiometric Technology Resource Centre
www.biometrics.orgwww.biometrics.orgUS Government focal research siteUS Government focal research site