Title :
A generalized biometric identification system model
Author :
Wayman, James L.
Author_Institution :
US Nat. Biometric Test Center, San Jose State Univ., CA, USA
Abstract :
Commonly-held knowledge and oft-repeated descriptions of biometric identification systems are more complicated than necessary because of the lack of a generally accepted system model. A good example is the supposed difference between "identification" and "verification". We present a system model applicable to all biometric systems showing the relationship between "verification" and "identification" and "enrollment" and "operation", illustrating where commonalities exist between seemingly disparate systems, and suggesting where interface standards might be useful. Further, we develop a general mathematical formulation to show "verification" to be the degenerate case of M-to-N matching, with N=1. Similarly, "PIN-less verification" is recognized as the degenerate case with small N with M=1, and "identification" can be seen as the more general case with small M and large N.
Keywords :
biometrics (access control); computer interfaces; identification; signal processing; telecommunication standards; M-to-N matching; PIN-less verification; data collection; decision subsystem; enrollment; error rate equations; general mathematical formulation; generalized biometric identification system model; interface standards; operation; signal processing; storage subsystem; transmission subsystem; Biomedical signal processing; Biometrics; Equations; Fingers; Iris; Multidimensional signal processing; Retina; Standards development; System testing; Taxonomy;
Conference_Titel :
Signals, Systems & Computers, 1997. Conference Record of the Thirty-First Asilomar Conference on
Conference_Location :
Pacific Grove, CA, USA
Print_ISBN :
0-8186-8316-3
DOI :
10.1109/ACSSC.1997.680201