A HAND VEIN STRUCTURE SIMULATION PLATFORM FOR ALGORITHM TESTING AND BIOMETRIC IDENTIFICATION

Septimiu Crisan, Ioan Gavril Tarnovan, Titus Eduard Crisan.

Abstract

Scanning the veins in human hands for biometric purposes has become more and more appealing due to inexpensive hardware requirements and rather simple software processing algorithms. However, there are few studies regarding the repeatability or uniqueness of the actual vein pattern. Furthermore, algorithm testing is usually carried out on a limited number of sample images acquired under different hardware setups. This is mainly due because of the lack of a vein pattern image database aimed to provide researchers with sufficient data to test recognition algorithms. Creating a large database of real hand vein patterns is a difficult task in terms of time and money. Other biometric domains, such as fingerprint recognition, benefit from synthetic generated images that greatly aid the accuracy of a given processing technique. This paper will offer a novel implementation of a platform for synthetic vein pattern images of the back of the hand. While the reasons behind pattern forming of veins are not completely known, there is sufficient data to create realistic images of vein patterns based on model reconstruction from crossing points, terminations and general anatomy of the hand.

Download

IMEKO-TC4-2008-101.pdf

DOI

-

IMEKO TC

TC4 - Measurement of Electrical Quantities

Back to the proceedings