Novel fuzzy vault method based on fingerprint characteristic data and matching algorithm

Abstract

The invention relates to a novel fuzzy vault method based on fingerprint characteristic data and a matching algorithm. A locking process in the method comprises the following steps: firstly, carrying out cyclic redundancy check (CRC) coding to random secret key data; then, constructing an associated polynomial; adding a farraginous point; carrying out a characteristic quantization process to each point in an assembly; and disorganizing the sequence of each point in the assembly; and generating a registered Hash table by the assembly. An unlocking process comprises the following steps: firstly, extracting input minutia characteristic data; then, quantizing each minutia characteristics data; generating a verification Hash table; comparing the verification Hash table with the registered Hash table by a matching algorithm to obtain a subset; carrying out a Lagrange interpolation reconstruction polynomial; and finally, carrying out a CRC verification process to the coefficient data of the polynomial obtained by reconstruction. According to the novel fuzzy vault method based on the fingerprint characteristic data and the matching algorithm, the secret key information and the fingerprint characteristic data are organically combined to effectively protect a secret key and simultaneously conceal the fingerprint characteristic template information of a user.

Description

technical field [0001] The invention belongs to the technical fields of pattern recognition and cryptography, and in particular relates to a novel fuzzy treasury method based on fingerprint characteristic data and a matching algorithm. Background technique [0002] In 2002, Juels and Sudan proposed the fuzzy vault scheme for the first time in their paper "A Fuzzy Vault Scheme", but there are still many problems that are not clearly explained. In 2003, Clancy et al. proposed a fingerprint fuzzy vault based on the work of Juels and Sudan. The concept of this scheme is the same as the original scheme of Juels et al. The biggest disadvantage is that it assumes that the fingerprint images have been pre-calibrated, but the actual situation is that it is impossible to obtain the query fingerprint image in advance for registration. In 2005, Uludag et al. proposed a more practical fuzzy vault algorithm, the most important idea of ​​which is to use the Helper Data extracted from the r...

AI Technical Summary

Problems solved by technology

[0002] In 2002, Juels and Sudan proposed the fuzzy vault scheme for the first time in their paper "A Fuzzy Vault Scheme", but there are still many problems that are not clearly explained. In 2003, Clancy et al. proposed a fingerprint fuzzy vault based on the work of Juels and Sudan. The concept of this scheme is the same as the original scheme of Juels et al. The biggest disadvantage is that it assumes that the fingerprint images have been pre-calibrated, but the actual situation is that it is impossible to obtain the query fingerprint image in advance for registration.

Most of the fingerprint template and key protection methods based on fuzzy vaults that have emerged in recent years are still at the stage of theoretical demonstration, and lack the necessary real and reliable experimental conditions.

Images