Rank-Level Fusion of Multimodal Biometric Cryptosystems

In this paper we present an effective scheme that preserves the biometric templates more secured. We present a new algorithm called Enhanced Blind Encryption Algorithm for securing biometric templates between client and server and using that for authentication. There are four phases involved in this process namely, enrollment, identification, matching and fusion. During enrollment, biometric traits (fingerprint and iris) will be uploaded and encrypted using Enhanced Blind Encryption Algorithm). During fusion, Borda count and Logistic regression methods are applied combinely in identifying the accurate results. Results indicate that the proposed system can be effectively used even for low-quality images. Keywords— multimodal, cryptosystem, rank-level fusion, fingerprint, iris, Borda count, EBEA