Abstract

Finger vein recognition systems utilize the venous pattern within the fingers to recognise subjects. It has been shown that the alignment of the acquired samples has a major impact on the recognition accuracy of such systems. Although a lot of work has been done to reduce the negative impact of finger misplacements, there is as yet no approach that solves all possible types of misplacement. In particular, longitudinal finger rotation still causes major problems. As the capturing devices evolve towards contactless acquisition, solutions for alignment problems are becoming increasingly important. As an alternative to rotation detection and correction, the problem can also be addressed by acquiring the vein pattern from different perspectives. This article presents a novel multi-camera finger vein recognition system which captures the vein pattern from multiple perspectives during enrolment and recognition. Contrary to existing multi-camera solutions, which use the same capturing device for enrolment and recognition, the capturing devices for the proposed system differ in the configuration of the acquired perspectives. The cameras of the devices are positioned in such way that the achieved recognition rate is high all around the finger and that the number of cameras needed is kept to a minimum. The experimental results confirm the rotation invariance of the proposed approach.

Reference

[Prommegger20b   ] A Fully Rotation Invariant Multi-Camera Finger Vein Recognition System Bernhard Prommegger, Andreas Uhl IET Biometrics 10:3, pp. 275-289, 2021

Data Set

PLUSVein Finger Rotation Data Set

The PLUSVein Finger Rotation Data Set (PLUSVein-FR) is a partly publically available finger vein data set. It contains finger images captured all around the finger from 63 different subjects, 4 fingers (index and middle finger from both hands) per subject, which sums up to a total of 252 unique fingers. Further information regarding the data set can be found by the following link:

PLUSVein-FR Data Set

Evaluation Framework Information

The experimental evaluations have been conducted using the open source vein recognition framework (PLUS OpenVein Finger- and Hand-Vein Toolkit) provided by the University of Salzburg. This is a feature extraction and matching/evaluation framework for finger- and hand-vein recognition implemented in MATLAB. It was tested on MATLAB 2016 and should work with all version of MATLAB newer or equal to 2016. This software is under the Simplified BSD license.

A more detailed description of the framework as well as its sources can be found here:

The framework contains all the feature extraction, comparison as well as evaluation methods used for the experiments in the paper.

Implementation: Deformable Finger Vein Recognition (DFVR)

The implementation of DFVR is based on the code of Deformable Spatial Pyramid Matching for Fast Dense Correspondences (external link) provided by Kim et al.. The code was extended by the vein based key-point selection PCA-Sift (external link) and bidirectional matching as described in the original DFVR paper (external link).

The DFVR implementation is available upon request. Please fill out this form to request a download link for the scores and settings files:

Requested Citation Acknowledgment

[Prommegger20b   ] A Fully Rotation Invariant Multi-Camera Finger Vein Recognition System Bernhard Prommegger, Andreas Uhl IET Biometrics 10:3, pp. 275-289, 2021

Implementation: Triplet-SqNet

The trained model of the used CNN approach Triplet-SqNet was taken from Finger Vein Recognition and Intra-Subject Similarity Evaluation of Finger Veins using the CNN Triplet Loss. It employes the SqueezeNet (external link) architecture using the triplet loss function together with hard triplet online selection (external link).

The download of the CNN model is available upon request. Please fill out this form to request a download link for the trained CNN model:

Requested Citation Acknowledgment

[Wimmer20a   ] Finger Vein Recognition and Intra-Subject Similarity Evaluation of Finger Veins using the CNN Triplet Loss Georg Wimmer, Bernhard Prommegger, Andreas Uhl In Proceedings of the 25th International Conference on Pattern Recognition (ICPR), pp. 400-406, 2020