Finger vein recognition

Description

During my master’s graduation assignment I explored the potential usage of the finger vein pattern as a biometric trait. All facets where covered during the assignment; sensor design, collection of a dataset, implementation and evaluation of several matching algorithms. The work was presented at the 2013 International Conference on Biometrics[1]. This initial sensor prototype led to several follow-up assignements which improved the sensor design even further.

Video

The below video shows the Miura et al. repeated line tracking algorithm[2] in action.

Dataset

The collected dataset, the University of Twente Finger Vascular Pattern (UTFVP) dataset, is available to the public after signing a license agreement.

References

1. B. Ton and R. N. J. Veldhuis, A high quality finger vascular pattern dataset collected using a custom designed capturing device 2013 International Conference on Biometrics (ICB), Madrid, Spain, 2013, pp. 1-5
2. Miura, N., Nagasaka, A. & Miyatake, T. Feature extraction of finger-vein patterns based on repeated line tracking and its application to personal identification. Machine Vision and Applications 15, 194–203 (2004)