5.7. Biometric Authentication Application Examples

Note that most biometric data can be adequately represented by a mixture of Gaussians, and also that RBF neural networks adopt basically the same type of Gaussian kernels. Therefore, RBF networks are more appealing for biometric applications than LBF networks. In fact, RBF and EBF networks have been used extensively in biometric applications. For instance, Brunelli and Poggio [38] used a special type of RBF network, called the "HyperBF" network, for face recognition and reported a 100% recognition rate on a 47-person database. In a speaker verification task, Mak and Kung [225] demonstrated that EBF networks perform substantially better than their RBF counterparts; and in a speaker identification ...

Get Biometric Authentication: A Machine Learning Approach now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.