Electronic Science and Technology ›› 2019, Vol. 32 ›› Issue (1): 32-37.doi: 10.16180/j.cnki.issn1007-7820.2019.01.007

Previous Articles     Next Articles

Finger Vein Authentication Technology Based on Image Registration

JIN Xiaofeng1,2,WANG Dandan2,LIANG Tiancai2,XU Jun2,WANG Biao1   

  1. 1. School of Physics,Sun Yat-Sen University,Guangzhou 510275,China
    2. Research Institute of GRG Banking Equipment Co., Ltd,Guangzhou 510663,China
  • Received:2017-12-25 Online:2019-01-15 Published:2018-12-29
  • Supported by:
    Pearl River Science and Technology Nova Project of Guangzhou, China(201710010152);Pearl River Science and Technology Nova Project of Guangzhou, China(201610010058);Science and Technology Planning Project of Guangdong Province, China(2016B010124005)


Venous distortion and image shadow that result from finger rotation and posture change during finger vein capture affect the accuracy of finger vein authentication. Therefore, this paper proposed a finger vein authentication method based on image registration, which authenticated the images by registering a few stable feature points, and extracting the texture features of the two registered images. In this method, it firstly selected the clues with high resolution in the neighborhood as the feature points, then matched the feature points and registered the two images. Finally, the similarity of the two registration images were calculated according to the texture features. The proposed method mainly overcame the problem of small number of points in the method based on feature point matching and solved the problem of venous deformation. Experimental results showed that this method had better performance than template matching and feature point matching.

Key words: finger vein authentication, feature point extraction, image registration, vein distortion, template matching, feature point matching

CLC Number: 

  • TP391.4