In biometric systems, people may be asked to provide multiple scans for redundancy and quality control. In the case of fingerprint matching systems, repeat fingerprint probes of the same physical finger can be available and data from such multiple samples can be fused for reliable authentication of individuals. Since multiple samples are from the same instance of the finger, some relationships between them, e.g. diversity or similarity, could be observed. In this paper, we investigate such relationships and use them in fusion in order to improve the performance of biometric systems. The relationships between samples are derived by measuring the similarity between matching score vectors with Pearson's correlation and cosine similarity measures. We conduct experiments using the FVC2002 dataset consisting of four fingerprint databases and trainable combination methods, likelihood ratio and multilayer perceptron. The results show that utilization of similarity measures for matching scores can further improve the multi-sample biometric fusion in both combination methods.
Mendeley saves you time finding and organizing research
Choose a citation style from the tabs below