Ranking Based Locality Sensitive Hashing Enabled Cancelable Biometrics: Index-of-Max Hashing

摘要

In this paper, we propose a ranking based locality sensitive hashing inspiredtwo-factor cancelable biometrics, dubbed "Index-of-Max" (IoM) hashing forbiometric template protection. With externally generated random parameters, IoMhashing transforms a real-valued biometric feature vector into discrete index(max ranked) hashed code. We demonstrate two realizations from IoM hashingnotion, namely Gaussian Random Projection based and Uniformly RandomPermutation based hashing schemes. The discrete indices representation natureof IoM hashed codes enjoy serveral merits. Firstly, IoM hashing empowers strongconcealment to the biometric information. This contributes to the solid groundof non-invertibility guarantee. Secondly, IoM hashing is insensitive to thefeatures magnitude, hence is more robust against biometric features variation.Thirdly, the magnitude-independence trait of IoM hashing makes the hash codesbeing scale-invariant, which is critical for matching and feature alignment.The experimental results demonstrate favorable accuracy performance onbenchmark FVC2002 and FVC2004 fingerprint databases. The analyses justify itsresilience to the existing and newly introduced security and privacy attacks aswell as satisfy the revocability and unlinkability criteria of cancelablebiometrics.