Cancelable multi-biometric security system based on double random phase encoding and cepstral analysis

Abstract

Biometric systems are widely used now for security applications. Two major problems are encountered in biometric systems: the security problem and the dependence on a single biometric for verification. The security problem arises from the utilization of the original biometrics in databases. So, if these databases are attacked, the biometrics are lost forever. Hence, there is a need to secure original biometrics by keeping them away from utilization in biometric databases. Cancelable biometrics is an emerging security trend in the field of biometric authentication. The objective of cancelable biometrics is to generate fake versions of the biometrics through non-invertible transforms or encryption methods to save the original biometrics from being compromised to guarantee their security. The other problem of biometric verification is the dependence on a single biometric, which reduces the trustiness of the verification results. Hence, there is a bad need to use multiple biometrics for trusted verification results. Multiple biometrics can be acquired for the same person and used for verification with a majority voting scenario to ensure trusted verification results. So, there is a need to save all biometrics in a secure way, which allows authentication from each of them, afterwards. The storage of multiple biometrics consumes storage space. Hence, there is a need for some sort of compression to save this storage space, while keeping the discrimination ability of subjects. This paper presents a novel approach that solves the security, trustiness, and storage problems of biometric systems. It is a cancelable multi-biometric security system based on Double Random Phase Encoding (DRPE) and cepstral analysis. Four biometrics are comprised in a unified biometric template for each person using Discrete Cosine Transform (DCT) compression. This unified biometric template is encrypted with the DRPE algorithm for security purposes. The cancelability is guaranteed through the ability to change the random phase sequences of the DRPE algorithm if the database is compromised. The multi-biometric compression is performed through keeping the most significant coefficients in the DCT domain for all four biometrics. The biometric recognition is performed by decrypting the unified biometric template and applying a cepstral approach for verification of the subject. A majority voting scheme can be followed for biometric verification at the receivers of remote-access biometric systems. The main advantage of the proposed cancelable multi-biometric system is the large degree of security, the immunity to communication channel effects through the utilization of a majority voting strategy at the receiver, the ability to withstand the compression effect, and the irreversibility through the implementation of cepstral features for biometric verification.