Method and device for finger vein recognition based on improved local binary model

Abstract

The invention discloses a finger vein recognition method and device based on an improved local binary model, which can overcome the influence of noise, improve the recognition accuracy and be easy to achieve in parallel. S1, acquire the finger vein image to be identified, locate the finger vein feature point position, and calculate the LmTP feature of each feature point based on the improved local binary model; S2, compare the finger vein by means of the fusion distance and the similarity of the LmTP feature Match the feature points with the finger vein feature points of the preset template image; S3, calculate the global similarity of the matching point pair according to the matching result; S4, perform finger vein recognition on the finger vein image according to the global similarity .

Description

technical field [0001] The invention relates to the fields of image processing and pattern recognition, in particular to a finger vein recognition method and device based on an improved local binary model. Background technique [0002] Vein recognition technology is Hitachi's discovery in the medical field that deoxygenated hemoglobin in veins can absorb near-infrared rays of a certain wavelength, and that human veins are unique. Finger vein recognition has attracted much attention as a second-generation biometric technology with high anti-counterfeiting. [0003] At present, the common finger vein recognition algorithm is a method using local information, such as LBP and its derived LDP, LTP, EQP and other methods. The method using local shape information first locates the feature points of the finger vein image, and then matches and recognizes the new finger vein image through the shape of the feature points. Because only the feature points are used, the template obtaine...

AI Technical Summary

Problems solved by technology

However, its disadvantage is that there are negative numbers in the singular vector obtained by SVD decomposition, which cannot explain its physical meaning well; on the other hand, the method of SVD decomposition is difficult to parallelize, resulting in a very slow speed

Huangrian's method, also known as the Hungarian algorithm, regards the matching problem as an allocation problem, but the algorithm is sometimes difficult to converge and cannot find the optimal solution

In addition, when these two methods match feature points, the probability of mismatching is relatively high.

Images