Score level fusion in multi-biometric identification based on zones of interest

In this paper, we present a new multibiometric fusion method for the identification of persons using two modalities, the iris and the fingerprint. Each modality is separately processed to generate a vector of scores. The fusion method is applied at the score level. A preliminary study based on the k-means clustering method, for each modality, led us to split the score range into three zones of interest relevant to the proposed identification method. The fusion is then applied to the extracted regions using two approaches. The first one achieves the classification by the decision tree combined to the weighted sum (BCC), while the second approach is based on the fuzzy logic (BFL). Several tests were conducted to evaluate the performance of the proposed methods on standard biometric databases using four metrics, namely, False Accept Rate, False Reject Rate, Enrollee False Accept Rate and Recognition Rate. The obtained results are very interesting since they illustrate clearly that the proposed fusion approaches outperform those based on a single modality. In addition, we showed that the BCC fusion approach achieves slightly better performance compared to the BFL. (c) 2019 The Authors. Production and hosting by Elsevier B.V. on behalf of King Saud University. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Automated summary

This paper presents a new multibiometric fusion method for individual identification using the iris and fingerprint modalities. Test results show that the proposed fusion methods outperform single modality methods, with the BCC fusion approach slightly outperforming BFL.