Multimodalb iometricsp rovidesh ighr ecognitiona ccuracy and populationc overageb yc ombining different biometrics ources.H owever,s ome multimodalb iometricsm ay obtains maller-than-expected improvemento f recognitionaccuracy if thecombinedbiometric sources aredependent in termsofa falsea cceptanceb ym istakenlyp erceiving biometricf eatures from twod ifferent persons as beingf romthe same person. In this paper, we proposeo ur multimodal biometricprototype that capturesapalm vein andthree

more »

... rprints simultaneously andw ee valuatew hether or nott heir combinationi ss tatistically independent. By evaluatingfalse acceptanceusing thepalmveinimages andthe fingerprintimages collected with our prototype, we confirmed that thec ombination of thep almv ein andthe fingerprints is almost independent.