While the various vendors of biometric products can throw out carefully polished statistics on their merchandise, so far there is very little comparison between the different types of products and their features. A study by International Biometric Group (IBG) takes a step toward filling that void.
The study, which was commissioned by several major financial organizations such as Visa, Chase Manhattan Bank, and Citibank, looks at eight finger scanning systems and two face geometry systems. It studies both how they perform and how users perceive them.
A significant factor in the study was that, instead of relying on engineers in a lab for product testing (the process used so far in most biometrics testing), subjects of the study were chosen from groups considered difficult to enroll in biometric systems. They were taken from three groups considered to cause finger-scanning problems: the elderly, who have delicate ridges; the Asian population, who have smaller ridges; and artists/construction workers, who have worn down ridges.
The online Electronic Identity Fraud Newsletter summarized some general findings that were significant. For example, one finding was that the two ways of typically measuring effectiveness - False Acceptance Rate (that is, how often an impostor can use someone else's identity) and False Rejection Rate (that is, how often a system incorrectly rejects a user), can have good ratings, but the system can still be ineffective if it has a third, largely unmeasured factor: that is, Failure to Enroll or how often people have difficulty enrolling to use the technology.
Another significant finding was that the different demographic segments of the population studied had different Failure to Enroll rates for different types of biometric products. This finding may be significant to financial institutions because when they choose between products, they must weigh the effectiveness of a product for use by their own customers' demographic groups.
A third interesting finding, according to the newsletter, was that background conditions such as bright lights affect both facial recognition software, and also finger-scanning equipment.
As far as how users perceive biometrics, the study set to rest a generally accepted theory that users connect finger printing with criminal activity. The study found that both finger and face technology was considered equally non-intrusive. Users did not equate electronic fingerprinting with the ink and roll systems used by law enforcement.
All the systems were considered easy to use by participants, but facial recognition systems were considered somewhat easier, the newsletter reports, and some fingerprint systems were considered harder than others.
For information on IBG, visit www.bio1.com. To receive a free subscription or for more information on Electronic Identity Fraud, visit www.e-dentification.com
BankersOnline is a free service made possible by the generous support of our
advertisers and sponsors. Please help us keep BankersOnline FREE to all
banking professionals. Support our advertisers and sponsors by clicking
through to learn more about their products and services.
Print Friendly!
Email This Article!
Discuss NOW!
