RSA cryptography based multi-modal biometric identification system for high-security application

Abstract

In recent days the requirements of Biometric Identification System (BIS) increased enormously. BIS Unimodal Biometric systems (UM-BS) have different kinds of problems like non-universality, noisy data, unacceptable error rate and spoof attacks. These limitations are solved by using multi-modal Biometric systems (MM-BS).MM-BS uses two or more individual modalities, like face, Palm, iris, retina, fingerprint, etc. This paper has introduced featurelevel fusion and Rivest Shamir Adleman (RSA) encryption based FEP-RSA-MM biometrics system. This FEP-RSAMM system has taken combination of Face, iris and Palm biological characters for individual Identification. FEPRSA- MM was implemented by using MATLAB and the performance were calculated and assessed in terms of Recall, Sensitivity, Specificity, Accuracy, F-Score, Precision, Mean Square Error, Root Mean Square (RMS) Error, etc. The performance of this FEP-RSA-MM system mainly depends on the accuracy. The accuracy of FEP-RSA-MM system is 93.33 % and it improved compared to two existing methods GF-FLF-MM, SIFT-KNN-MM, FLF-GSO-MM and SLF-PSO-MM.