Finger Vein Recognition Method Based on Convolutional Neural Network for Deformation Detection and Correction

Abstract

Finger vein recognition method based on convolutional neural network deformation detection and correction, aiming at the problem that the existing finger vein recognition algorithm has poor recognition performance when the finger is severely deformed, this method carefully analyzes the principle of finger vein deformation correction and The convolutional neural network is introduced into the finger vein correction and recognition algorithm, and the recognition model that meets the performance requirements is trained as much as possible in the case of a small data set. The method can accurately correct the deformation of the finger vein image and extract effective features for identification when the finger is irregularly deformed. The invention fully considers the process of finger vein deformation correction, uses a convolutional neural network for finger vein deformation correction, and uses an efficient feature extraction network to extract and identify finger vein features. At the same time, the original finger vein image is directly used as the input of the method, which does not require complex preprocessing operations, and can achieve good recognition results in small-scale data sets.

Description

technical field [0001] The invention relates to a method for performing deformation detection and correction on finger vein images and identifying veins, belonging to the field of human biological features. Background technique [0002] Biometrics is a security technology that uses the unique biometrics of the human body for identification. Biometrics refer to physiological characteristics or behaviors that are unique (distinguished from others), universal (normal people have), stability (no mutation), and descriptiveness (can be extracted and stored), such as fingerprints , palm print, iris, veins, face, gait, EEG, etc. Compared with traditional identity authentication technology, biometric technology has been increasingly valued by various research institutions and even countries due to its characteristics of being safer, more convenient, and more stable, and has been widely used in various industries. [0003] Existing finger vein recognition systems are roughly divided...

AI Technical Summary

Problems solved by technology

Aiming at the problem that the existing finger vein recognition algorithm has poor recognition performance when the finger is severely deformed, this method carefully analyzes the principle of finger vein deformation correction and introduces convolutional neural network into the finger vein correction and recognition algorithm. In the case of a small data set, train a recognition model that meets the performance requirements as much as possible

Images