Fingerprint storage and comparison method based on behavior analysis

Abstract

The invention discloses a fingerprint storage and comparison method based on behavior analysis and relates to the field of fingerprint comparison. The method comprises the following steps of adding a time frequency label and a spatial position label to every fingerprint template in a fingerprint database based on behavior characteristics of a user, grading the fingerprint templates according to time frequency and spatial positions, dividing comparison processes into three grades including local comparison, network comparison and server comparison, wherein the local comparison, the network comparison and the server comparison respectively correspond to the most common local fingerprint databases, common network fingerprint databases and non-common server fingerprint databases, the comparison process of every time firstly occurs in the local fingerprint database, and obtaining a result if the local comparison is successful; performing the network comparison if the local comparison fails, performing distributed comparison by using the fingerprint databases in within a certain regional range, and obtaining the result if the network comparison is successful; and performing the server comparison if the network comparison fails by using the non-common server fingerprint databases stored in a server. According to the fingerprint storage and comparison method based on the behavior analysis disclosed by the invention, the comparison range can be effectively reduced, the comparison time can be shortened, and the comparison efficiency can be improved.

Description

technical field [0001] The invention relates to the field of fingerprint comparison, in particular to a method for fingerprint storage comparison based on behavior analysis. Background technique [0002] Biometric authentication technology is a technology that uses human biological characteristics to determine personal identity, and can be widely used in criminal investigation, security, banking and other fields. In current applications, fingerprint identification and DNA identification technologies are relatively mature, among which fingerprints are the most commonly used human biometrics due to their characteristics of easy collection and easy comparison of features, and fingerprint identification has quickly occupied most of the biometric technology market. As a biometric technology, fingerprint comparison has been widely used in many fields. To judge whether a fingerprint is legal, it is necessary to compare the feature value of the fingerprint to be compared with the f...

AI Technical Summary

Problems solved by technology

[0003] However, with the increase in the number of fingerprint users, the fingerprint library gradually increases. In large-scale fingerprint storage and comparison applications, especially when the fingerprint library reaches tens of thousands or even millions of levels, it is necessary to select from N fingerprint templates Find out the fingerprints that need to be authenticated. N is a natural number greater than 1. The 1:N comparison of fingerprints will put forward very high requirements on equipment, algorithms, and time.

In order to shorten the fingerprint comparison time, the traditional method is: bind a code to the fingerprint, and compare the codes first, find the fingerprints corresponding to the codes, and then compare them, so that the 1:N The comparison is converted into a 1:1 comparison, or a clustered large-scale computer is used for comparison. Although these can shorten the time of fingerprint comparison to a certain extent, they do not fundamentally solve the problem. In large-scale comparisons, it is still not easy to deal with

When a fingerprint user outputs a fingerprint, the traditional method needs to submit the fingerprint to a large background fingerprint database. According to the shape of the database, centralized comparison or distributed comparison may be used, so that each comparison must use 1: The way of N, when N is large, the efficiency is very low and it takes a long time

Images