Fingerprint image representation method based on a variety of feature points

Abstract

The invention provides a fingerprint image representation method based on a variety of feature points, which is applied to small-area fingerprint image recognition and belongs to the technical field of automatic fingerprint recognition. The method comprises the steps of image preprocessing, feature point extracting and feature point matching. In view of the defect that a small-area fingerprint contains less information, an image representation method based on a variety of fingerprint feature points is put forward. Fingerprint feature points are divided into two categories, wherein the center point, bifurcation points and endpoints of a fingerprint are classified to a first category, and ridge line points near the bifurcation points are classified to a second category. Then, fingerprint matching is performed by making comprehensive use of a variety of information of fingerprint feature points. By using the method put forward by the invention, the number of fingerprint feature points and the amount of useful fingerprint feature information are increased, and the accuracy of small-area fingerprint recognition can be improved greatly.

Description

technical field [0001] The invention relates to an automatic fingerprint recognition technology, in particular to a small-area fingerprint automatic recognition method, especially a fingerprint image with a small area and less information. Background technique [0002] At present, automatic fingerprint identification technology is the most widely used biometric identification technology, and its identification accuracy and operational feasibility are higher than other identification methods. The service of replacing application passwords with fingerprint authentication has made automatic fingerprint recognition technology popular in the market. [0003] However, the accuracy rate of the current automatic fingerprint identification algorithm is far lower than that of the expected fingerprint identification products on the market, and both the identification accuracy rate and the identification speed need to be improved. Especially for small-area fingerprint recognition, the ...

AI Technical Summary

Problems solved by technology

[0003] However, the accuracy rate of the current automatic fingerprint identification algorithm is far lower than the expected accuracy rate of fingerprint identification products on the market, and the identification accuracy rate and identification speed need to be improved

Especially for small-area fingerprint recognition, the amount of information provided by the image is reduced, the recognition accuracy is reduced, the security of use is reduced, and the difficulty of fingerprint recognition is increased.

Images