Revocable palm print feature generation method and system based on minimum signature

Abstract

The invention discloses a revocable palm print feature generation method and system based on a minimum signature. The method comprises the following steps of obtaining an original palm print image, extracting a palm print region of interest (ROI) in the original palm print image, and extracting the orthogonal features of the palm print ROI; randomly generating a chaotic matrix as a secret key, andperforming XOR calculation on the chaotic matrix and the orthogonal matrix to obtain an initial characteristic matrix; and randomly generating a plurality of hash functions to generate a first signature matrix, carrying out scanning calculation on the initial feature matrix according to the hash functions, and replacing the maximum value in the signature matrix with the minimum value to obtain asecond signature matrix serving as the revocable palm print feature; and finally, carrying out scanning calculation on the initial feature matrix by utilizing irreversibility of a Hash function, and generating a final minimum Hash signature matrix as a final revocable palm print feature. According to the present invention, the security and the privacy of palm print biological characteristics are effectively protected, and the security and the privacy can be improved under the condition that the recognition rate is ensured.

Description

technical field [0001] The present application relates to the technical field of palmprint recognition, in particular to a method and system for generating revocable palmprint features based on minimum signatures. Background technique [0002] With the enhancement of people's security awareness, traditional verification methods based on keys, ID cards, passwords, etc. can no longer meet people's security needs, and traditional identity authentication methods are prone to risks such as loss, forgetting, and theft. With the development of information technology, the most convenient and safe authentication method is biometric identification. At present, the widely used identification methods include face recognition, fingerprint recognition, iris recognition and palmprint recognition. Compared with several other biometric identification methods, palmprint recognition has the disadvantages that the collection equipment is cheap, it is not easily affected by light, and palmprint ...

AI Technical Summary

Problems solved by technology

[0002] With the enhancement of people's security awareness, traditional verification methods based on keys, ID cards, passwords, etc. can no longer meet people's security needs, and traditional identity authentication methods are prone to risks such as loss, forgetting, and theft.

Images