An Introduction To

An Introduction to BIOMETRICS

Agenda • Introduction To Biometrics • Definition Of Biometrics • Mode Of Biometrics • Type Of Biometrics • Fingerprint Recognition • Application

How Can I Know Who You Are ?

How Can You Prove it ?

Identifying Individuals There are three common ways of identifying an individual's identity •Token ---- What the individual has An Identification Card, Badge, Magnetic Strip Card

•Secret/Knowledge ---- What the individual Knows/Owns A Password or Personal Identification Number (PIN)

•Biometrics ---- What the individual is or does

Biometrics …….. The Definition “The Automated reorganization of individuals based on their behavioral and Biological characteristics.”

The term “ Biometrics” is derived from the Greek words Bio ( Means ‘Life’ ) and Metric (means ‘To Measure’ )

Biometrics …….. The Truth • Universal : Each person should have the specific Biometric character.

• Unique (Just Distinguishable) : Any two people should be sufficiently different in terms of the characteristics.

• Collectible : Biometric characters must be obtainable.

• Permanent : Characters remain invariable over time, allowing for repeatable measures.

How It Works

Type Of Biometrics • Biological Face Finger Hand Iris Vein Retina Imaging  Ear  Odor      

• Behavioral    

Signature Keystroke Voice Gait

SECURITY AT YOUR FINGERTIP

Fingerprint Recognition

Fingerprint recognition is a Biometric technique which use different characteristics of a person’s fingerprint.

The use of fingerprints as a means of identification were first used by the Chinese in 500 B.C.

Details About Fingerprint

Tool Used

Finger Token is essentially a personal password storage tool with all the capabilities required for fingerprint authentication—scanning fingerprints, storing fingerprint reference data, and matching the verification.

Digital Fingerprint Any image data can be converted into digital information with the combination of “0” and “1”. In this case, even though we convert a fingerprint into digital data, the property of uniqueness of a fingerprint is still valid. < Conversion without loss>

Eg.

00011101101010000111101

> <

Conversion with allowing loss > 0

1

1

0

0

1

1

Fig..Conversion of fingerprint to Binary data

The second row of binary fingerprint is created by taking every third bit of the first row of full conversion of original fingerprint. Therefore a fingerprint can be converted into unlimited number of new digital fingerprint.

Pattern Matching Have a digitalized target data & digitalize the fingerprint ; now we should enter pattern-mapping phase To map the pattern of the fingerprint into the data, we use a brute-force matching between the fingerprint and the data. Starting from the first bit of fingerprint, step by step find the maximum length of bits of fingerprint that is matching with the content bits of the data.

Eg. For eg: there are two sequence of binary information one is for fingerprint and other for data.

Overall Process Pattern Encoding

Fingerprint Scanning

1001 1111 0001 1101

Mapping+Fingerprint Pattern

Mapping

01010101 01110110 0101010 01110100 0111011 01010111 0111010 0101011

Create Data

0100110 0111011 0001010 0101011

010101 011101 011101 010101

Distribution to other users

1001 1111 0001 1101

Extract Fingerprint

Applications Forensics Information system/computer network security Electronic commerce and electronic banking Fingerprint-based national ID systems Fingerprint-based voter and driver registration

Others Face Recognition Based on location and shape of facial attributes such as the eyes, eyebrows, nose, lips and chin, as well as their relationship to each other

Hand Recognition Based on measurements of the hand

Iris Recognition Iris is the area between the pupil and the sclera. Iris carries very distinctive information. 1 in (10)78 – chance of someone having an identical iris as you.

Others Vein Recognition •Finger based •Top of hand based •Palm

Keystroke Recognition How an individual types on a keyboard

Voice Recognition •Text dependent Voice recognition system is based on utterances of a fixed predetermined phrase • Text independent Voice recognition system recognizes the speaker independent of what they say

Keystroke press time

Keystroke latency time

Pros & Cons • Pros:

• Cons

 The uniqueness of biometric attributes

Still very expensive

No question of forgetting passwords More difficult to replicate a biometric feature at the data acquisition stage

The hardware and software must be integrated properly System may not recognize a valid user Biometric authentication gives a likelihood of a match

Fusion

When more than one biometric is combined, fusion occurs. Multi-modal

Conclusion Most security breaches are due to weak authentication…….

Is Biometrics the answer ?

Thank you