A portable multimodal biometric system

Roger Ouch, University of Louisville
Roman Yampolskiy, University of Louisville

Abstract

Many multimodal biometric systems have been proposed during the past decade, but these systems have been deployed in laboratories and often lack of interoperability and portability. In this paper, we propose a multimodal biometrics system using face and voice biometrics implemented on a Mac OSX laptop using built-in sensors. We use standard programming language and libraries to achieve adaptability and portability on varied platform. Our bimodal biometric system includes hardware-dependent and hardware-independent package to achieve maximum flexibility and adaptability. The results showed that our bimodal biometric achieved good accuracy rates, portability and adaptability to protect personal computers. Experiments are also encouraging and provide opportunity for future development and deployment in real-life applications to secure personal computers, laptops or smartphones.