Towards a Measure of Biometric Information

Abstract

This paper addresses the issue of the information content of a biometric image or system. We define biometric information as the decrease in uncertainty about the identity of a person due to a set of biometric measurements. We then show that the biometric information for a person may be calculated by the relative entropy D(pparq) between the population feature distribution q and the person's feature distribution p. The biometric information for a system is the mean D(pparq) for all persons in the population. In order to practically measure D(pparq) with limited data samples, we introduce an algorithm which regularizes a Gaussian model of the feature covariances. An example of this method is shown for PCA and ICA based face recognition, with biometric information calculated to be 45.0 bits (PCA), 39.0 bits (ICA) and 46.9 bits (fusion of PCA and ICA features). Finally, we discuss general applications of this measure