Dynamic random projection for biometric template protection

Random projection provides a good diversification effect for biometric template protection but is drawing increasing security concerns under the token-stolen (or public parameter) case. We propose a dynamic random projection method to alleviate these security concerns due to the stolen token by increasing the computational complexity to search for the unprotected biometric features. This is achieved by a projection process which dynamically assembles a random projection matrix from a set of candidate projection vectors. The selection of projection vectors is decided by the biometric feature vector itself and thus forms a nonlinear projection process. The proposed method permits the public and secure storage of all the candidate random vectors without the need for external secret keys. Experiments on the 800 samples in the database FVC2002DB2_A demonstrate the well-kept biométrie performance of the proposed method.