Professor Harry Wechsler
Department of Computer Science
e-mail : wechsler@cs.gmu.edu
web : http://cs.gmu.edu/~wechsler/
(703) 993-1533 (office)
(703) 993-1530 (sec)
(703)993-1710 (fax)
SPRING
'2006
CS 667 Biometrics
Class Information
001 10912
R (“Thursday”) 7:20 p.m. –
10:00 p.m. Robinson A249
Prerequisites
CS 580 or permission of the instructor
Office Hours
Thursday 6:15 – 7:00 PM or by appointment (SITE II -
Rm. 461)
Textbook
Guide to Biometrics
by Bolle, Connell, Pankanti,
Ratha, and Senior, Springer 2004.
References
1. Biometric Identification in Networked
Society by Jain, Bolle and Pankanti (Eds.), Kluwer, 1999.
2. Biometric Authentication by Kung, Kak, and Lin,
Prentice Hall, 2005.
Course Description
Basic
principles and methods for automatic
authentication of individuals. Technologies include
face,
fingerprint and iris recognition,
and speaker verification. Additional topics cover
multimodal biometric
and data fusions, system design, performance evaluation, and privacy
issues. Term project
required.
Motivation
Biometrics, the science of recovering or verifying a
person's identity, measures the physical or behavioral characteristics that
make people unique—including fingerprints, an eye's retina or iris, face, hand
geometry, signature and voice—and uses those measurements for personal
authentication. Biometrics is related to the science of forensics, which uses
and interprets physical evidence for legal purposes. The importance of biometrics lies in the fact
that traditional means of identification and verification are often unreliable
or cumbersome: Passwords are difficult to remember and easy to steal. Keys,
driver's licenses, and passports can be lost or forged. The human body and its
behavior, on the other hand, can't be forgotten, stolen, forged, or misplaced.
Practical uses for biometrics are wide spread and include maintaining the
security for both physical and cyber space. In particular, biometrics aids in
controlling access to an office, computer network or an ATM, smart cards,
wireless communication; confirming the identity of buyers and sellers to make
electronic commerce safe and reliable; confirming student identity for distant
learning; and safeguarding electronic records related to health care services.
Emerging trends in biometrics include data fusion, augmented cognition, and error analysis. Human behavior, a subject of interest for W5+ (who, where, when, what, why, and HOW) and closely related to Human-Computer Intelligent Interaction (HCII), expands on biometrics and improves performance. The scope for biometrics is multi- and inter- disciplinary as it draws from several fields, ranging from signal and image processing, computer vision and pattern recognition, speech processing, machine learning, to cognitive and neurosciences.
Follow – up
Graduate Certificate in Biometrics http://cs.gmu.edu/programs/compbiocertificate.html
CS 775 / IT 844 – Pattern Recognition – Spring ‘2007
Doctoral Dissertation
Grading
Homework à 15%
Midterm à Thursday, March 23 à 20%
(Team) Term Project à April 27 and May 4 à 40%.
Final à Thursday,, May 11 à 25 %
Topics
System
Design.
Sensing and Data Collection.
Subspace Methods for Representation.
Predictive Learning.
Authentication and Identification.
Detection and Tracking.
Data Fusion.
Augmented Cognition.
Performance Evaluation
and Error Analysis.
Security and Privacy.