Biometric Identification Science and Technology

In this course you will learn:

• How to classify your application.
• Which biometric methods might be most applicable to your problem.
• The fundamentals of fingerprinting, iris scanning, speaker verification, hand geometry, and dynamic signature recognition technologies.
• Statistical testing and analysis methods.
• General one and two-dimensional transform theory.
• Pattern recognition techniques.
• The design and testing of a speaker verification system.
• Performance prediction of Large-scale identification systems
• The legal, social and ethical concerns of applying these technologies.

Course Description

This course is designed to cover Biometric Identification Science and Technology with a balance between the basic theoretical background and practical application.

We will discuss:

• Taxonomies of Devices and Applications
• Probability and Statistical Testing Methods
• One and Two Dimensional Transform techniques
• Device Specifics: finger printing, voice recognition, facial recognition, and iris scanning
• Large Scale Identification Applications
• Social, Legal, and Ethical Concerns

Cost per course through SJSU Open University: $450 for ENGR 297.

Maximum 6 units of SJSU Open University courses can be credited toward the Master of Science degree.

Instructor Dr. James L. Wayman, Director of the U.S. National Biometric Test Center located in the San Jose State University College of Engineering, is an internationally recognized expert on Biometric Identification Systems.

SPECIAL TOPICS ENGINEERING 297:
INTRODUCTION TO AUTOMATIC BIOMETRIC IDENTIFICATION PREREQUISITES: STATISTICS, CALCULUS, C PROGRAMING COURSE PROJECT: WRITE AND TEST A GENDER RECOGNITION SPEECH PROCESSING ALGORITHM

SYLLABUS
WEEK 1
INTRODUCTION

TERMINOLOGY
TAXONOMY OF DEVICES
TAXONOMY OF APPLICATIONS
PATTERN RECOGNITION BASICS

WEEK 2
MATHEMATICAL UNDERPINNINGS

VECTOR SPACES
DISTANCE MEASURES
NON-EUCLIDEAN SPACES
COMPLEX SPACES
DISTANCE DISTRIBUTIONS
CORRELATION MATCHING

WEEK 3
BINOMIAL TESTING

PROBABILITY FUNDAMENTALS
DERIVATION OF BINOMIAL DISTRIBUTION
POISSON APPROXIMATION
CONFIDENCE INTERVALS
TEST SIZE

WEEK 4
PROBABILITY

BAYES RULE
MULTIPLE MEASURES
ONE-TO-N, ONE-TO-MANY, M-TO-N
AND, OR, XOR
GENUINE, IMPOSTOR, TEMPLATE DISTRIBUTIONS

WEEK 5
TRANSFORM METHODS

TRANSFORM CONCEPTS
ONE-D EXAMPLES
GENERALIZED FOURIER
WAVELET
HARTLEY
CEPSTRAL
TWO-D EXAMPLES
FOURIER
WAVELET

WEEK 6
SPEAKER RECOGNITION

CEPSTRAL X-FORMS
MEL-SCALE
STANDARDIZED CORPRA
COHORT TESTING

WEEK 7
HAND GEOMETRY

CASE STUDY: INSPASS
DRUNKS, LAMPPOSTS AND FFT CONVOLUTION
CASE STUDY: SJSU COMP. CENTER
SJSU DATABASE

WEEK 8
FACIAL RECOGNITION

1888 GALTON PAPER
ROCKEFELLER U WORK
MIT WORK
FERRET DATABASE
THERMOGRAPHY

WEEK 9
FINGERPRINTING: FORENSIC

ANSI/FBI STANDARDS
IMAGE QUALITY
WSQ COMPRESSION
DATA FORMAT
ANSI BENCHMARK TEST PLAN

WEEK 10
FINGERPRINTING:

NON-FORENSIC STANDARDS
PHILIPPINE BENCHMARK TEST PLAN
CASE STUDY: LA AFIRM SYSTEM
LARGE-SCALE SYSTEMS
PHASE-IN PROBLEMS
COST-BENEFIT ANALYSIS
COMPUTATIONAL SPEED LIMITS
CENTRALIZED VS. NON-CENTRALIZED SYSTEMS

WEEK 11
EYE METHODS

RETINAL
IRIS

WEEK 12
PROJECT PRESENTATIONS
FINAL EXAM