Multimodal Biometric Authentication with Secured Templates — A Review

Abstract

The advent of internet is highly triggering people towards the digital era. The inexpensive data availability to common people also aids for the abundant data communication. In the meantime, securing data and applications from unauthorized identities is becoming very crucial. Biometric authentication system utilizes physiological and behavioral characteristics of individuals to automate access control and identity verification. These fast, convenient and un-obstructive systems can be categorized as unimodal and multimodal, based on the number of biometric traits or modalities used. Multimodal biometric systems have shown improved success rate of authentication with increased robustness over unimodal biometrics at the cost of facing numerous complications. The unauthorized access to biometric templates which represents distinctiveness of people is one of the major threats to user's privacy and security. Hence, variety of template protection techniques can be practiced for securing biometric templates. This paper gives an overview of multiple biometrics used for authentication. Focusing on challenges in multimodal biometrics practiced through various fusion levels and different algorithms, this paper discusses scope of further research in this field. Majorly, the template security issue of multimodal biometrics is emphasized with various techniques to protect the crucial asset of human identity.