Finger vein recognition method combining apparent similarity and singular point matching number

Abstract

The invention provides a finger vein recognition method and system combining apparent similarity and singular point matching number. The method comprises the following steps: preprocessing each image in a data set for finger vein recognition; obtaining features of the test image and the template image based on the spatial overlapping pyramid and orthogonal local preserving projection dimensionality reduction, and calculating the apparent similarity between the test image and the template image based on the features; calculating the singular point matching number of the test image and the template image, and calculating the singular point matching number similarity between the test image and the template image based on the singular point matching number; performing weighted fusion on the apparent similarity and the singular point number similarity of the test image and the template image to obtain a weighted matching score; and classifying the weighted matching score values of the test image and each template image, wherein an image label corresponding to a classification result is arecognition result. The problem that the recognition rate is low due to different finger postures during finger vein collection can be effectively solved.

Description

technical field [0001] The invention relates to the technical field of finger vein recognition, in particular to a finger vein recognition method and system combining apparent similarity and singular point matching numbers. Background technique [0002] At present, the methods for finger vein recognition are mainly divided into two categories: deep learning methods and traditional methods. The deep learning methods mainly include self-learning feature extraction methods based on neural networks. This method is based on the learning of massive finger vein image data. Under the constraints of the frame theory, adjust the parameters of the multi-layer network, establish the best nonlinear fitting network between the input and output, and complete the finger vein recognition with the goal of image classification. The traditional method is based on feature extraction, mainly based on texture feature extraction method, by extracting the vein network from the gray image of the fing...

AI Technical Summary

Problems solved by technology

[0004] In view of this, the present invention provides a finger vein recognition method and system that combines the apparent similarity and the number of singular point matches, which can effectively solve the problem of low recognition rate due to different finger postures during finger vein collection

Images