Dynamic Score Selection for Fusion of Multiple Biometric Matchers

Abstract

A biometric system for user authentication produces a matching score representing the degree of similarity of the input biometry with the set of templates for that user. If the score is greater than a prefixed threshold, then the user is accepted, otherwise the user is rejected. Typically the performance are evaluated in terms of the receiver operating characteristic (ROC) curve, and the equal error rate (EER). In order to increase the reliability of authentication through biometrics, the combination of different biometric systems is currently investigated by researchers. While a number of "fusion" algorithms have been proposed in the literature, in this paper we propose a theoretical analysis of a novel approach based on the "dynamic selection" of matching scores. Such a selector aims at choosing, for each user to be authenticated, just one of the scores produced by the different biometric systems available. We show that the "dynamic selection" of matching scores can provide a better ROC than those of individual biometric systems. Reported results on the FVC2004 dataset confirm the theoretical analysis, and show that the proposed "dynamic selection" approach is more effective when low quality scores are used

HMMPayl: An intrusion detection system based on Hidden Markov Models

“…It is easy to see that the distribution of the output values of the Ideal Score Selector allows a better separation between the classes with respect to each of the combined classifiers. It can be easily seen that the above Ideal Score Selector exhibits a better ROC curve than the ROC curves of each individual classifiers used in the combination, and consequently a larger AUC [14]. Moreover, it has also been shown that the Ideal Score Selector always attains a larger AUC than that obtained by the linear combination, whatever the value of the weights, and the number of classifiers [14].…”

“…The "Ideal Score Selector", proposed in [14], represents the upper bound of the selection strategy. The output of such Ideal Score Selector can be computed as:…”

“…Besides, the MCS paradigm guarantees that the weaknesses of classifiers due to a sub-optimal choice of initial parameters are mitigated. In the same step, we also evaluated the performance of a classifiers selection approach that provides an upper bound on accuracy and may suggest some guidelines to exploit the combination of classifiers [14]. We deeply tested HMMPayl on several datasets of legitimate traffic and attacks.…”

Enhancing Performance of Multibiometric System using Ant Colony Optimization based on Score Level Fusion

“…In this section we only mention the related works of score level fusion which is based on evolutionary technique like Pso, Ga, Aco and other optimization technique. Dynamic selection of matching score was proposed by Tronci et al [8] in which they show that the dynamic selection of matching score can provide a better performance than a unimodal system. Veeramachaneni et al [9] was proposed an adaptive multimodal biometric management algorithm which is based on the combination of PSO and Bayesian decision fusion.…”

An Adaptive Multimodal Biometrics System using PSO

“…Tronci et al [8] have recently investigated another aspect of multimodal problem that focused on the dynamic selection of matching scores from all the available matching scores. The best matching score from a set of matching scores was selected based on the likelihood of input user being genuine or impostor.…”