scispace - formally typeset
Search or ask a question

Showing papers on "Biometrics published in 1991"


Patent
20 Jun 1991
TL;DR: In this paper, a continuous biometric authentication matrix is proposed, which can employ any number of any variety of biometric personal identification devices including thumbscan, digital photo, voiceprints, fingerprints, and so on.
Abstract: A system activating and analyzing the biometric data from a plurality of biometrically-oriented personal identification devices at intermittent intervals selectively allows or prevents continued use of a particular protected system or device by a particular individual. The system is a continuous biometric authentication matrix which can employ any number of any variety of biometric personal identification devices including thumbscan, digital photo, voiceprints, fingerprints, and so on. The system acts as a continuously functioning "gate" between a system to be protected and a prospective user. A prospective user's biometric data is stored for reference. When the prospective user attempts to use the protected system or device, he or she must interface with the system, which then compares the prospective user's biometric data to the reference data. This comparison must not only be acceptably close in similarity in order to gain access to the protected system, it must also continue to be close in subsequent comparisons in order for access to the protected system or device to continue. The accept/reject threshold for individual biometric sensor devices is adjustable, as is the accept/reject threshold for the overall combination of biometric sensors.

414 citations



MonographDOI
01 Jan 1991
TL;DR: This paper summarizes the methodology for fitting loglinear models to three-dimensional contingency tables using the method of maximum likelihood and outUnes the logistic regression model for a dichotomous dependent variable.
Abstract: This paper contains two parts. The first part summarizes the methodology for fitting loglinear models to three-dimensional contingency tables using the method of maximum likelihood. An example based on traffic accident data is used to illustrate the techniques. The second part of the paper outUnes the logistic regression model for a dichotomous dependent variable. An example based on female labor force paxticipation data is used to demonstrate the methodology.