Biometrics

What is Biometrics?
 Automated method for recognizing individuals
based on measurable biological and behavioral
characteristics
Finger Print Recognition
 Minutiae

 Pattern Matching

 Problems: sometimes
unusable
Vascular Pattern Matching
 LED infrared light

 Fingers and back of

hand

 Not completely viable

Iris Recognition
 Uses infrared light

 Converts Images to
vectors

 Needs further
development
Facial Recognition
 Location and position
of facial features

 Dependent on
background and
lighting conditions
Voice Verification
 Factors: pitch, intensity,
quality and duration

 Text dependent

 Text independent

 Problems: include
background noise
Hand Geometry
 Scan both sides of hand

 Primarily used for

verification

 Not as accurate as other

methods
Dynamic Signature
 Factors: velocity,
acceleration and speed

 Mainly used for

verification

 Problems: forgers
could reproduce
Retina Recognition
 One of the most secure
means of biometrics

 Unique to each person

 Unique to each eye

 Problems: require effort

on the part of subjects
Other Types
 Keystroke

 Gait

 DNA

 Odor
Commercial Applications
 Computer login

 Electronic Payment

 ATMs

 Record Protection
Government Applications
 Passport control

 Border control

 Access Control
Forensic Applications
 Missing Persons

 Corpse identification

 Criminal investigations
Type of Authentication
 Authentication
 1:1

 Verification
 1:N
Constraints on Biometrics
 Typical “Constrained”
Image
 Constraints:
 Lighting
 Distance
 Pose
 Expression
 Time Lapse
 Occlusion
Constraints on Biometrics
 “Unconstrained” Image
Biometrics Research at Clemson
 Biometric and Pattern
Recognition Lab

 Goals:

1. Usable Biometrics
2. Unconstrained
Biometrics
Biometrics Research at Clemson
 Aging Research
Biometrics Research at Clemson
 Demographics
 Older vs. Younger
 Males vs. Females
 Geographic origin of
algorithms
Biometrics Research at Clemson
 PeriocularRegion
Recognition
 Texture, color, eye
shape
 Overcome facial
occlusion
Biometrics Research at Clemson
 Ear Recognition
 Not affected by aging
or expression
 Covert collection of
images
 Little research
performed
Sources
 Biometrics.gov. Web. 05 Dec. 2011.
http://www.biometrics.gov/ReferenceRoom/Introduction.aspx
 Jain, Anil K., Arun Ross, and Salil Prabhakar. "An Introduction to Biometric
Recognition." IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR
VIDEO TECHNOLOGY 14.1 (2004): 4-20. IEEE Xplore. Web. 5 Dec. 2011.
http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1262027
 Jain, Anil K., Patrick J. Flinn, and Arun A. Ross. Handbook of Biometrics. New
York: Springer. Web. 5 Dec. 2011. http://libcat.clemson.edu/record=b2478857
 Phillips, Jonathon P., Alvin Martin, C. L. Wilson, and Mark Przybocki. "An
Introduction Evaluating Biometric Systems." Computer 33.2 (2000): 56-63. IEEE
Xplore. Web. 5 Dec. 2011.
http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=820040
 http://bprl.cs.clemson.edu/about.html
 http://bprl.cs.clemson.edu/projects.html