Designed to automatically recognize individuals based on biological and behavioral traits – such as fingerprints, palm prints, or voice/facial recognition, biometric systems are “inherently fallible,” according to a new report by the National Research Council. According to the report, no single trait has been identified that is stable and distinctive all groups, and additional research is needed at virtually all levels of design and operation in order to strengthen the science and improve system effectiveness.
“For nearly 50 years, the promise of biometrics has outpaced the application of the technology,” says Joseph Pato, chair of the committee that wrote the report and technologist at Hewlett-Packard’s HP Laboratories in Palo Alto, CA. “While biometric systems can be effective for specific tasks, they are not nearly as infallible as their depiction in popular culture might suggest.”
According to the report, the systems provide “probabilistic results,” which means that confidence in results must be tempered by an understanding of the inherent uncertainty in any given system. When the likelihood of an imposter is rare, even systems with accurate sensors and matching capabilities can have a high false-alarm rate, which can become costly or dangerous in systems designed to provide heightened security.
The report states that careful consideration is needed when using biometric recognition as a component of overall security system. Any biometric system selected for security purposes should undergo thorough threat assessment to determine its vulnerabilities to deliberate attacks. The report outlines several features that a biometric system should contain.
Copies of Biometric Recognition: Challenges and Opportunities are available from the National Academies Press. Visit www.nap.edu.