Biometrics

DR. SHERYL BRAHNAM CIS DEPARTMENT MISSOURI STATE UNIVERSITY

Who are you?

Who are you?

Identifying fellow human beings has been crucial to the fabric of human society There have been times and places where communities were small enough that everyone knew each other
With population growth and increasing

mobility, documents and secrets were needed to establish identity

Who are you?

Person identification is now an integral part of the infrastructure needed for Law enforcement Government agencies Diverse business sectors

How Can People Be Indentified?

People are identified by three basic means: Something they HAVE (Signet-rings, Letter of Introduction)
Something

they KNOW (Baseball teams in NYC)

Something

they ARE (Man with six fingers)

Automatic Identification
Possession based

(something you HAVE)

Credit card Smart card

Knowledge-based

(something you KNOW)

Password PIN

Biometric-based

(something you ARE)

Finger prints Facial appearance

Problems with Possession & Knowledge-Based Approaches

Cards may be lost or stolen or forgotten Passwords are forgotten Passwords are not secure Passwords and pins can be guessed 25% of the population write their pins on their ATM card Traditional approaches are unable to

differentiate between an authorized person and an imposter

Business Costs
The average adult working in a

large business has 12 passwords to remember And spends nearly a week in every year logging into systems.

Business Costs

The average cost to a large company for every password lost is $16.

Black Market Identity Information

Now a billion $ annual

industry (see The Hacker Economy, InformationWeek, February 12, 2007) While theft goes up, use goes down

10.2 10 9.8 9.6 9.4 9.2 9 8.8 8.6 8.4 8.2 2003 2005 2006 2005 9.3 million $54.4 billion total loss $5,885 average loss 2003 10.1 million $53.2 billion total loss $5,249 average loss

2006 8.9 million $56.6 billion total loss $6,383 average loss

Black Market Identity Information

Over 250 million U.S. identities

compromised in last 3 years But use of identities is going down Identity information is being stock-piled-bought and sold as an asset on the Internet Your identity today is the same 5 years from now

Black Market Identity Information

$490 credit card with PIN $147 drivers license $98 social security number $6 PayPal account logon and password $78-$294 billing data including account

number, address, social security number, home address, and birth date
Data from Trend Micro

Why Biometrics?

Why Biometrics?

Biometrics

What are Biometrics?

Biometrics The automated recognition of

individuals based on biological and behavioral characteristics

(SC 37 Working Group Proposed Definition)

What are Biometrics?

Biometry mathematical and statistical

analysis of biological data Biometric system a pattern recognition system that recognizes a person by determining the authenticity of a specific biological and/or behavioral characteristic (biometric)

What are Biometrics?

Anthropometrymeasurement techniques of

human body and its specific parts

Forensic (judicial) anthropometryidentification of criminals by these measurement techniques

Requirements for an Ideal Biometric Identifier

1. Universality Every person should have the biometric characteristic 2. Uniqueness No two persons should be the same in terms of the biometric characteristic 3. Permanence The biometric characteristic should be invariant over time

Requirements for an Ideal Biometric Identifier

4. Collectability The biometric characteristic should be measurable with some (practical) sensing device 5. Acceptability One would want to minimize the objections of the users to the measuring/collection of the biometric 6. Ethics Even if acceptable, is the collection and use of the biometric ethical?

Biometrics Systems

ENROLLMENT VERIFICATION RECOGNITION

Enrollment
Enrollment

Person entered into the database

Verification
Verification: Am I who I claim to be? One to one comparison

Identification
Identification: Who am I?

One to many comparison

Identify, Verify, & Vulnerability

Verify:

Determine if identity claim is true Identify: Determine if the person is already enrolled in the system Vulnerability: Susceptibility of the machine to being tricked:

Convince it that I am you Convince it that I am not me

Why are there errors?

The Challenges: How to guarantee that the sensed measurements are not fraudulent How to accurately represent and recognize biometric patterns? How to acquire repeatable and distinctive patterns from a broad population

Using Different Biometrics

Possible Biometrics
Biological traces

DNA (DeoxyriboNucleic Acid), blood, saliva, etc. Biological (physiological) characteristics fingerprints, eye irises and retinas, hand palms and geometry, and facial geometry

Possible Biometrics
Behavioral characteristics

dynamic signature, gait, keystroke dynamics, lip motion Combined voice

Popular Biometric Characteristics

DNA: The Ultimate Biometric

One-dimensional unique code for ones individuality, but identical twins have identical DNA patterns

DNA: The Ultimate Biometric

Issues limiting the utility of DNA Contamination Access Automatic real-time recognition issues Privacy issues: information about susceptibilities of a person to certain diseases could be gained from the DNA pattern

Behavioral vs Physical Traits

Physiological

Behavioral

Eye Fingerprints Hand Shape and Veins Face

Gait Keystrokes Signature Voice

On-line and Off-line Signature

Dynamic Signature

Fingerprints

Fingerprints: A Brief History

1750 B.C. - In ancient

Babylon, fingerprints were used on clay tablets for business transactions. 300 B.C. - Emperors of China used personalized clay seals 220 A.D. - The first time ink prints were used was in china

Fingerprints: A Brief History

1858 - Sir William Herschel, Chief Magistrate

of the Hooghly district in Jungipoor, India, first used fingerprints on native contracts. 1880 - Dr. Henry Faulds published an article in the Scientific Journal, "Nautre" (nature) discussed:

fingerprints as a means of personal identification, use of printers ink as a way of obtaining them.

Fingerprints: A Brief History

1892 - Sir Francis Galton published his book,

"Fingerprints

Establish the individuality and permanence of fingerprints. Included the first classification system for fingerprints.

Fingerprints (Uses)

Fingerprint Systems
Thermic Systems (heat) Capacitive Systems (electrical charge)

Fingerprint Features

Hand Shape

Palm

Hand and Vein Recognition

Hand and Vein Recognition

Iris Recognition: Eye

Iris

Iris Code

Comparing Irises
Different IrisCodes are

compared (ExclusiveORing) Occluded (nonvisible) parts of the iris are masked.

National Geographic 1984 and 2002

Retina
Every eye has its own totally unique pattern of blood vessels

Face Recognition

Facial Recognition Systems

Face Recognition: Correlation

Problems: Lighting Variability

Indoor
Outdoor

Same person or different person?

Same person or different person?

Same person or different person?

Other Problems

Pose Facial hair Occlusion Age

Facial Recognition System

Face Classification: PCA

Principal Component Analysis Developed to classify identity

M. A. Turk and A. P. Pentland, "Face recognition using eigenfaces," presented at IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Silver Spring, Md, 1991.

Face Classification: PCA

Successful at classifying Gender

A. J. O'Toole, H. Abdi, K. A. Deffenbacher, and J. C. Bartlett, "Classifying faces by race and sex using an autoassociative memory trained for recognition," presented at Proceedings of the 13th Annual Conference on Cognitive Science, 1991.

Face Classification: PCA

Developed to classify identity Successful at classifying
Gender

(OToole and Deffenbacher)

Age

D. Valentin, H. Abdi, A. J. O'Toole, and G. W. Cottrell, "Connectionist models of face processing: A survey," Pattern Recognition, vol. 27, pp. 1209-30, 1994.

Face Classification: PCA

Successful at classifying
Gender

(OToole and Deffenbacher) Age (Valentin, et al.) Facial Expression

G. W. Cottrell and J. Metcalfe, "EMPATH: Face, emotion, and gender recognition using holons," in Advances in neural information processing systems, D. Touretzky, Ed. San Mateo: Morgan & Kaufman, 1991, pp. 564-71.

Face Classification: PCA

Image space and face cluster

PCA Process

Original Images Average Face

Face Classification: PCA

The Eigenface Generation Process

Original Images

Eigenvectors

PCA Process
Eigenvectors (they are face like)

Face Classification: PCA

A face images can be represented exactly as weighted combinations of the eigenface components
= * W1 * W1
*W 1

Facial Recognition Systems

Thus a face is a series of weights or points in

a multidimensional space

Facial Recognition Systems

To classify a face, find its position in face

space and find the closes point (face) to it. If it is close enough, call it a match

Advantages of Face Recognition

Voluntary Action vs Passive Usage Data Acquisition

- 5% cannot provide good Fingerprint Cost - Cheaper than Iris Detection (movement)

Comparison of Biometric Techniques

Comparison of Biometric Techniques

Problems Using Biometrics

Intra-class variability

Inter-class similarity

Identical Twins

Temporal Variations

Segmentation: Face Detection

Locating Faces in a Crowd

Noisy Images

More Problems
Different types of fraud Loopholes: illegitimate access not envisaged

by the designers

More Problems
Inherent in the means and mechanisms of

identification used by the system that cannot be completely eliminated

Collusion; misuse of super operator status Coercion; forcibly extract biometric from a user Denial; user denies access after gaining the access Covert Acquisition; video snooping

Biometric Applications

Applications

UCSD Biometric Soda Machine

Access Control

Did You Vote?

Video Surveillance (Airports)

Fingerprint System at Gas Stations

Galp Energia SGPS SA of

Lisbon won the technology innovation award for developing a payment system in which gasoline-station customers can settle their bills simply by pressing a thumb against a glass pad. Scanning technology identifies the thumbprint and sends the customer's identification information into Galp's backoffices system for payment authorization.
THE WALL STREET JOURNAL, November 15, 2004

Iris Scans to Unlock Hotel Rooms

The Nine Zero hotel in Boston just installed a new system which uses digital photos of the irises of employees, vendors and VIP guests to admit them to certain areas, the same system used in high-security areas at airports such as New York's JFK.

Fingerprint Systems at Border Crossing

New Passports

Want to charge it?

Then talk to your credit card

Biometrics for personalization

Application Characteristics
Overt vs Covert Attended vs Unattended Cooperative vs Non-cooperative

Scalable vs Non-scalable
Private vs Public Reduce vs Eliminate fraud

(Ethical) Questions?

Biometrics and BI
International Biometric Group

(www.biometricgroup.com) How Stuff Works (www.howstuffworks.com)