Neil Yager

The Biometric Menagerie

It is commonly accepted that users of a biometric system may have differing degrees of accuracy within the system. Some people may have trouble authenticating, while others may be particularly vulnerable to impersonation. Goats, wolves, and lambs are labels commonly applied to these problem users. These user types are defined in terms of verification performance when users are matched against themselves (goats) or when matched against others (lambs and wolves). The relationship between a users genuine and impostor match results suggests four new user groups: worms, doves, chameleons, and phantoms. We establish formal definitions for these animals and a statistical test for their existence. A thorough investigation is conducted using a broad range of biometric modalities, including 2D and 3D faces, fingerprints, iris, speech, and keystroke dynamics. Patterns that emerge from the results expose novel, important, and encouraging insights into the nature of biometric match results. A new framework for the evaluation of biometric systems based on the biometric menagerie, as opposed to collective statistics, is proposed.

Worms, Chameleons, Phantoms and Doves: New Additions to the Biometric Menagerie

Goat, sheep, and lamb are labels for problem users of a biometric system. These user types are defined in terms of their verification performance when matched against themselves (goats) or when matched against others (lambs and wolves). Four new members of the biometric menagerie are proposed based on a users relationship between their genuine and imposter match scores. Experiments using a variety of biometrics, algorithms, and data sets have indicated that this relationship can exist with real-world data (face, voice, and fingerprint). An analysis of the results indicates that the presence of these user types tends to reflect a weakness of the underlying system. A new framework for the evaluation of biometric systems based on user groups from the biometric menagerie (as opposed to collective statistics) is proposed.