A Zero-Leakage Fuzzy Embedder From the Theoretical Formulation to Real Data

摘要: In this paper, we present a novel biometric cryptosystem obtaining perfect security, that is not leaking any information about the employed secret key from knowledge of stored helper data. While similar purposes have already been sought in literature, approaches proposed so far evaluated terms recognition performance under unrealistic assumption ideal statistical distributions for considered Conversely, investigate applicability framework to practical scenarios while managing trade-off between privacy and performance. This goal has achieved by introducing class transformation functions enforcing zero-leakage secrecy, designing an adaptive strategy embedding bits into selected features, developing system parameters optimization with respect performance, privacy. Experimental tests conducted on real fingerprint data prove effectiveness scheme.