Minutiae-Based Structures for A Fuzzy Vault

Abstract

One vital application of biometrics is to supplement or replace passwords to provide secure authentication. Cryptographic schemes using passwords require exactly the same password at enrolment and verification to authenticate successfully. The inherent variation in samples of the same biometric makes it difficult to replace passwords directly with biometrics in a cryptographic scheme. The fuzzy vault is an innovative cryptographic construct that uses error correction techniques to compensate for biometric variation. Our research is directed to methods of realizing the fuzzy vault for the fingerprint biometric using minutia points described in a translation and rotation invariant manner. We investigate three different minutia representation methods, which are translation and rotation invariant. We study their robustness and determine their suitability to be incorporated in a fuzzy vault construct. We finally show that one of our three chosen structures shows promise for incorporation into a fuzzy vault scheme.