OPH-based template protection method for revocable binary features

Abstract

The invention discloses an OPH-based template protection method for revocable binary features. The method comprises the steps: calculating a first fixed length vector according to the similarity between a to-be-registered fingerprint image and an MCC descriptor of each sample fingerprint image in a sample fingerprint database in response to a fingerprint registration instruction; mapping the first fixed-length vector into a first aggregation vector; performing matrix multiplication on the first aggregation vector and a local Haar matrix, and repeating a product result for n times to obtain a first expansion vector; generating a random replacement seed with the same length as the first expansion vector, and performing position replacement on elements in the first expansion vector by using the random replacement seed to obtain a first replacement vector; and uniformly segmenting the first replacement vector, and generating a binary feature as a first fingerprint protection template according to the position of the minimum element in each segmented vector. According to the method for extracting the fingerprint features, original biological features do not need to be stored, a large amount of storage space is not occupied, and the matching speed is high.

Description

technical field

[0001] The invention belongs to the technical field of fingerprint templates, in particular to a template protection method based on OPH (One permutationhashing, one permutation hashing) revocable binary features. Background technique

[0002] Compared with traditional password authentication, fingerprint authentication has the advantage that users do not need to worry about password forgetting or password leakage. However, the current fingerprint authentication systems all store the user's original fingerprint features. There is a loophole in this approach, that is, once the stored original fingerprint features are leaked, it will cause large-scale irreparable losses. Moreover, since most of the original fingerprint features are images, the fingerprint authentication system needs to store a large number of original fingerprint features of users on the server side, which will occupy a large amount of storage space on the server; in addition, the matching ver...

Images