Enhancement of Multimodal Biometric Systems Using an Improved Unconstrained Cohort Normalisation

(Fawaz Assa'edi)

---

Abstract

An important factor affecting the performance of multimodal biometrics in practice is that of contamination in the individual biometric data types. This paper proposes an improved score normalisation technique to reduce the effects of such contamination in the multi-modal fusion. It is shown in this paper that by deploying such technique at the score level, the adverse effects of data degradation can be reduced considerably. The experimental investigations involve the two recognition modes of verification and open-set identification in mixed-quality data conditions. The paper presents the motivation for, and the potential advantages of, the proposed approach and details the experimental study. Keywords: Multimodal biometrics; Score-level fusion; Unconstrained cohort normalisation; Biometric verification.
PDF

References