A Finger Vein Image Recognition Method Based on Locally Symmetrical Graph Structure

Abstract

A Finger Vein Image Recognition Method Based on Locally Symmetrical Graph Structure. It includes using Gabor filtering method to perform multi-channel filtering on finger vein ROI images to obtain multi-directional Gabor filtering images; constructing a feature encoding graph structure based on a local symmetric graph structure; performing feature encoding in the 0° direction on the feature encoding graph structure; The feature coding map structure in the ° direction is rotated counterclockwise by 45°, 90° and 135° respectively to obtain multi-directional feature coding values; select the maximum value of the feature coding values ​​as the single-channel feature coding value of the central pixel point, and obtain the other 7 The respective single-channel feature encoding maps of the channels are composed of all 8-channel single-channel feature encoding values ​​to form the feature encoding image of the 8-channel filtered image; steps such as finger vein ROI image similarity matching are performed. The invention fully expresses the local position information and gradient information between neighborhoods, and is successfully applied to vein image recognition, thus improving the recognition performance.

Description

technical field [0001] The invention belongs to the technical field of finger vein image recognition, in particular to a finger vein image recognition method based on a local symmetrical graph structure. Background technique [0002] With the development of biometric identification technology, vein identification has been widely concerned as a new identification method because of its natural anti-counterfeiting property. Compared with traditional biometric patterns (such as fingerprints, faces, iris and palm prints), finger vein patterns have certain advantages in uniqueness, versatility, durability, and high security. At present, finger vein recognition has been widely used in various fields, including computer login, security check, ATM authentication, etc. [0003] Finger vein images can obtain internal vein image features through near-infrared light transmission or reflection fingers, effectively avoiding obstacles to accurate image features caused by factors such as wr...

AI Technical Summary

Problems solved by technology

Because the traditional finger vein feature expression method is sensitive to illumination changes and skin scattering problems, and for the problems caused by the variable finger posture, the traditional finger vein feature expression method cannot effectively solve the problem of finger vein feature expression due to the limitation of rotation invariance. The problem of variable posture

Images