Biometrics experts have long assumed that the iris is a stable identifier throughout a person’s lifetime. New research from the University of Notre Dame challenges that concept. Researchers discovered that the iris ages and subtly changes over time, making recognition more difficult and less accurate over time Biometric systems compare subjects’ irises with prior readings of approved users’ irises. If they match, a person is granted access to whatever the system is controlling. If an authorized person’s iris does change over time, scans won’t match stored readings and access will be denied. “[I]t was believed from the early days of iris recognition that there was no template aging effect, nobody bothered to look for the effect,” stated Notre Dame professor Kevin W. Bowyer. “Also, only recently have research groups had access to [biometric retina] image datasets acquired for the same people over a period of several years.” Despite these findings, Bowyer said iris-recognition security should still be effective by taking steps such as having users take new iris readings periodically, an approach known in the field as re-enrollment, so that the system will continue to recognize them as their iris changes. The researchers presented their work at the recent IEEE Computer Society Workshop on Biometrics. (PhysOrg)(University of Notre Dame)(Paper – Kevin W. Bowyer Website)
