Home

“Identity Crisis” (biometrics), Business New Haven, October 29, 2001

Identity Crisis Biometrics posts major advances as a post-9/11 security tool
By: Lisa Micali	Business New Haven

Since the September 11 World Trade Center and Pentagon attacks, biometrics has made front-page headlines. Defined as a unique, measurable physical or behavioral characteristic of a human being for automatically recognizing or verifying identity, biometrics characteristics can include fingerprints, iris data, hand and face geometry, signature, voice and DNA. Each of these methods has different degrees of accuracy, cost, social acceptability and intrusiveness.

Businesses around the world have taken part in a wide range of pilot projects involving biometrics. In Germany, facial recognition is being incorporated into an ATM system by Siemans Nixdorf. Iris recognition is being incorporated into Japanese ATMs by that country's leading supplier, Oki Electronic. MasterCard and VISA are exploring the use of biometrics to increase the security of their credit cards (annual losses to credit-card fraud are estimated at $2 billion - much of which, it is believed, could be eliminated through the use of biometrics). Customers of the Standard Bank of South Africa are having their fingerprints scanned at ATM machines instead of using a PIN (personal identification number) when they wish to withdraw cash, make deposits or inquire about their balances.

Chemical Bank recently announced its introduction of a voice-verification system for customer identification in banking transactions. Charles Schwab & Co. uses a finger-scanning system with employees in its main office to conform to requirements of the U.S. Securities & Exchange Commission. Other businesses, such as Coca-Cola, have incorporated biometrics into time and attendance systems.

In Connecticut, the state's Department of Social Services (DSS) uses biometrics to assist in the delivery of various entitlement programs. The state legislature passed a law requiring that and Aid to Families with Dependent Children (AFDC) and General Assistance (GA) recipients be biometrically imaged for identification purposes. The DSS Digital Imaging System became operational on January 1, 1996.

David Mintie is program manager for DSS' Biometric Identification Project. He is responsible for technology acquisition, contract implementation and ongoing contract management. He also has responsibility for the state's Electronic Benefit Transfer (EBT) program, in addition to directing the Central Processing Division of DSS. Mintie is also editor of the Biometrics in Human Services User Group newsletter (BHSUG). He currently chairs a national group of human-services biometric program directors attempting to establish standards for the interstate exchange of finger-imaging data with the goal of improving program integrity for human services agencies.

Give us a brief history of your project in Connecticut and why the state is using it.

Our implementation strategy was to work together with other agencies to create a standard hardware platform that would serve the needs of both the Department of Social Services (DSS) and the Department of Motor Vehicles for image-capture and ID card production. DSS implemented a statewide electronic personal identification system that accurately enrolls qualified General Assistance (GA) and Aid to Families with Dependent Children (AFDC) clients into a statewide database. We issue tamper-resistant identification cards that incorporate finger-image “identifiers” stored in two-dimensional bar codes. Uses of the finger-image identification verify that enrolled clients are eligible to receive benefits. Connecticut's main objective was to create a system that would eliminate dual enrollments' or what we call 'double-dipping' [welfare fraud]. Our objectives included biometric identification/verification based on finger-imaging technology; centralized image-storage and -retrieval; and over-the-counter identification card production.

How does the program work?

Digital images are created for every new and existing welfare recipient. These images are stored in a computer database along with a digitally captured facial portrait and signature. As each new applicant is imaged, the digital record is matched against the established database in real time. Connecticut has selected a combination of proven technologies that places its system at the cutting edge of all human services systems in place in the U.S. It's also compatible with other systems to catch 'double-dippers' from other states.

What technology do you use?

The equipment used in the digital-imaging process includes a Pentium III computer, an LCD signature tablet, a small optical fingerprint reader, a PVC card printer and a digital camera. Applicants place their two index fingers, one at a time, on the fingerprint scanner. Applicants can see their own fingerprints on the computer screen while the computer scans their fingerprints into the central database. While their fingerprints are being recorded and matched, the system operator will take their photograph and record the applicant's signature. In less than five minutes, a real-time match process is completed and the applicant is given a tamper-proof, secure photo identification card. The card contains the applicant's photo, welfare identification number, a 2-D bar-code containing fingerprint minutiae data for fast one-to-one identification verification, and a ISO standard magnetic stripe that can carry everything from EBT financial transaction codes for use in ATMs and POS devices to medical-eligibility data for medical service providers.

Have you ever caught anyone from out of state?

Yes. We matched images from New Jersey and New York and caught several people. However, we didn't feel it was cost-effective to perform these interstate matches. We are currently looking at ways to make it more cost-effective.

How many people have you caught since 1996?

We feel it is an extremely effective deterrent and have identified 108 cases to date that we have referred to our fraud division. You may say, That's not a lot. But the real benefit of the system is its use as a deterrent. A good [parallel] I like to use is a burglar-alarm system. Imagine that you ask people how many burglars they've caught. Most people would say none, or wouldn't be able to tell you. The signs on the front lawn [for the system] serve as a deterrent in pretty much the same way as our biometric cards. It's effective. It's secure and it prevents criminal activity.

What about privacy concerns?

The safeguards in place are legislative. The data is fraud-proof and the information gathered through the digital-imaging process is subject to our confidentiality rules and may not be used for purposes other than program administration. As I stated earlier, interstate matches with other human-services agencies are permissible.