Finger Vein Recognition Based On Non-negative Sparse Matrix Decomposition

The rapid development of technological advances have brought great convenience to the society, and with the help of internet information exchanges quickly and widely. At the same time, information security and privacy is particularly important, and people have higher requirements for identity authentication. Due to inherent physiological or behavioral characteristics,biometric identification technology has great advantages and is widely used. At present, finger vein recognition technology as a new member in the field of biometric identification,overcomes many shortcomings of fingerprint recognition, hand shape recognition and other hand features, has aroused extensive attention of many researchers.This paper research on finger vein recognition, including image processing and feature extraction algorithm, and design a complete set of finger vein recognition system.First, change the finger-vein image from RGB to gray and do normalization operations to reduce storage space. The full finger regions are extracted by using Otsu threshold method based on the column direction and getting rid of the small block noise through connected domain area. In order to reduce the influence of nonlinear translation and rotation when using finger vein images obtained with non-contact devices, we propose a region of interest extraction method. First through fitting finger midline to identify the deflection angle and rotation correction;then, using the diameter of the fingertip contour to locate, find the finger specified width maximum inscribed rectangle and finally size normalization. Comparative experiments are done between complete finger-vein region and finger-vein region of interest,the results show this algorithm can accurately extract the region of interest, and effectively improve the system recognition performance.Subspace method is commonly used in feature extraction, data dimensionality reduction via matrix factorization to find the algebraic properties of the data. Non-negative matrix factorization is a new decomposition, adding non-negative constraint in the decomposition process and making the decomposition results also satisfy non-negative. Therefore, this paper study using non-negative matrix decomposition to extract local characteristics of the finger vein. First study the traditional non-negative matrix factorization theory, then research two improved sparse algorithm, and analysis for finger vein characteristics extracted. The article contrast the three algorithms from the feature-based image, the convergence rate, verify accuracy and identification accuracy, and draw many conclusions.As a multi-resolution analysis method, wavelet decomposition can reflect the image at different scales in different directions, this paper studies the theory of two-dimensional wavelet transform and the wavelet decomposition subgraph significance, and propose a non-negative sparse matrix factorization based on wavelet transform and analyzing the influence of wavelet decomposition level and wavelet bases on the recognition rate. The approximation image filters off high frequency noise and contains almost all the information of the original image. Matrix factorization of the this subgraph and adjustment sparse factor of feature matrix and coefficient matrix to extract vein characteristics. This can reduce storage space and the computational complexity. The matching distance distribution curve shows that the feature has a good distinguish ability.Finally, build finger vein recognition system to verify the identification of the various algorithms. Experimental results show that the proposed algorithm not only ensure a fast convergence rate, but also significantly shorten the training time, recognition rate also increase slightly.