Information Technology Reference
In-Depth Information
A Virtual Curtain for the Detection of Humans
and Access Control
Olivier Barnich, Sébastien Piérard, and Marc Van Droogenbroeck
INTELSIG Laboratory, Montefiore Institute, University of Liège, Belgium
Abstract. Biometrics has become a popular field for the development
of techniques that aim at recognizing humans based upon one or more
intrinsic physical or behavioral traits. In particular, many solutions ded-
icated to access control integrate biometric features like fingerprinting or
face recognition.
This paper describes a new method designed to interpret what happens
when crossing an invisible vertical plane, called
hereafter,
at the footstep of a door frame. It relies on the use of two laser scanners
located in the upper corners of the frame, and on the classification of the
time series of the information provided by the scanners after registration.
The technique is trained and tested on a set of sequences representative for
multiple scenarios of normal crossings by a single person and for tentatives
to fool the system.
We present the details of the technique and discuss classification re-
sults. It appears that the technique is capable to recognize many scenarios
which may lead to the development of new commercial applications.
virtual curtain
1
Introduction
Detecting a person, locating him, and recognizing its identity are three corner-
stones of applications turned on security. Over the past years, many technologies
based on biometrical signatures have emerged to achieve these goals. The hand-
book by Jain et al. [1] illustrates the many techniques available today. They
ranges from fingerprinting, voice recognition, face detection, dental identifica-
tion techniques to iris, gesture or gait recognition, just to name a few.
In this paper, we propose a new platform (comprising hardware and soft-
ware) for critical applications such as secure access control, where biometrics
has became a viable technology, that can be integrated in identity management
systems. Commonly, access to restricted areas is monitored by a door with an
electrical lock or a revolving door activated by the swipe of an access control
card. In this context, we aim for a system able to send an alarm when the ex-
pected scenario of a single person crossing the door frame is not confirmed; this
could occur when someone enters a restricted area by passing through the door
at the same time as another person (this is called piggybacking when the other
person is authorized and tailgating when the other person is unauthorized).
The purpose of our method is to identify the scenario of one or more persons
when they cross a door frame. While camera driven solutions exist for it, we
 
Search WWH ::




Custom Search