Biometric score fusion in identification model using the Choquet integral

In this paper, we present a novel method for multibiometric identification system based on fusion at matching score level using the Choquet integral and the associated fuzzy measure. The output of each biometric matcher is modeled as a fuzzy set using membership functions, and the corresponding fuzzy entropy estimates the reliability of the information provided by each biometric matcher. Then, the fuzzy densities are generated based on training accuracy and fuzzy entropy. The results are aggregated using the Choquet integral. This novel method can measure the importance of each biometric matcher and improves the accuracy of the fusion model significantly.