SHREE VAISHNAV INSTITUTE OF

TECHNOLOGY AND SCIENCE

BIOMETRICS SYSTEMS

Guided By: Submitted By:

Mr. SAURABH PANDEY VINEET KUMAR KHOTI (1115)

E.C.(B)
4th YEAR
CONTENTS

• Introduction to biometric
authentication

• Biometric methods

• Biometric system architecture

• Biometric Application
  What is user authentication?

The process of confirming an individual’s identity,

either by verification or by identification

 A person recognising a person

 Access control (PC, ATM, mobile
phone)
 Physical access control (house, building,
area)
 Identification (passport, driving licence)

 Authentication methods

• Token – “something that you have”

• such as smart card, magnetic card, key,
passport, USB token
• Knowledge – “something that you know”
• such as password, PIN
• Biometrics – “something that you are”
• A physiological characteristic (such as
fingerprint, iris pattern, form of hand)
• A behavioural characteristic (such as the
way you sign, the way you speak)
  What is biometrics?

• The term is derived from the Greek words bio

(= life) and metric (= to measure)
• Biometrics is the measurement and statistical
analysis of biological data
• In IT, biometrics refers to technologies for
measuring and analysing human body
characteristics for authentication purposes
• Definition by Biometrics Consortium –
automatically recognising a person using
distinguishing traits

 How does it work?

• Each person is unique

• What are the distinguishing traits that make
each person unique?
• How can these traits be measured?
• How different are the measurements of these
distinguishing traits for different people?
 Verification vs. identification

• Verification (one-to-one comparison) –

confirms a claimed identity
• Claim identity using name, user id, …
• Identification (one-to-many comparison) –
establishes the identity of a subject from a set
of enrolled persons
• Employee of a company?
• Member of a club?
• Criminal in forensics database?

 Biometric identifiers

• Universality
• Uniqueness
• Stability
• Collectability
• Performance
• Acceptability
• Forge resistance
  Biometric technologies
• Fingerprint biometrics – fingerprint
recognition
• Eye biometrics – iris and retinal scanning
• Face biometrics – face recognition using
visible or infrared light (called facial
thermography)
• Hand geometry biometrics – also finger
geometry
• Signature biometrics – signature
recognition
• Voice biometrics – speaker recognition

Other biometric methods

• Found in the literature:

 Vein recognition (hand)
 Palmprint
 Gait recognition
 Body odour measurements
 Ear shape
 DNA
 Keystroke dynamics
 Static vs. dynamic biometric methods

• Static (also called physiological) biometric

methods – authentication based on a feature
that is always present
• Dynamic (also called behavioural) biometric
methods – authentication based on a certain
behaviour pattern

 Classification of biometric methods

 Static
 Fingerprint r.
 Retinal scan
 Iris scan
 Hand geometry

 Dynamic
 Signature r.
 Speaker r.
 Keystroke dynamics
 Biometric system architecture

• Major components of a biometric system:

 Data collection
 Signal processing
 Matching
 Decision
 Storage
 Transmission

 Biometric system model

Raw Extracted
data features Template
Signal Matching Storage
processing

Match
score

Decisio
Application n
Authentication decision
Data collection subsystem

• Also called data acquisition

• Comprises input device or sensor that reads the
biometric information from the user
• Converts biometric information into a suitable
form for processing by the remainder of the
biometric system
• Examples: video camera, fingerprint scanner,
digital tablet, microphone, etc.

Requirements for data collection

• Sampled biometric characteristic must be

similar to the user’s enrolled template
• The users may require training
• Adaptation of the user’s template or re-
enrolment may be necessary to accommodate
changes in physiological characteristics
• Sensors must be similar, so that biometric
features are measured consistently at other
sensors

Changes in data collection

• The biometric feature may change
• The presentation of the biometric feature at the
sensor may change
• The performance of the sensor itself may
change
Signal processing subsystem

• For feature extraction

• Receives raw biometric data from the data
collection subsystem
• Transforms the data into the form required by
matching subsystem
• Discriminating features extracted from the raw
biometric data
• Filtering may be applied to remove noise

Matching subsystem

• Key role in the biometric system

• Receives processed biometric data from signal
processing subsystem and biometric template
from storage subsystem
• Measures the similarity of the claimant’s
sample with the reference template
• Typical methods: distance metrics,
probabilistic measures, neural networks, etc.
• The result is a number known as match score

Decision subsystem
• Interprets the match score from the matching
subsystem
• A threshold is defined. If the score is above the
threshold, the user is authenticated. If it is
below, the user is rejected
• Typically a binary decision: yes or no
• May require more than one submitted samples
to reach a decision: 1 out of 3
• May reject a legitimate claimant or accept an
impostor

Storage subsystem

• Maintains the templates for enrolled users

• One or more templates for each user
• The templates may be stored in:
 physically protected storage within the
biometric device
 conventional database
 portable tokens, such as a smartcard

Transmission subsystem

• Subsystems are logically separate

• Some subsystems may be physically integrated
• Usually, there are separate physical entities in
a biometric system
• Biometric data has to be transmitted between
the different physical entities

Biometric data is vulnerable during transmission

Enrolment

• Process through which the user’s identity is

bound with biometric template data
• Involves data collection and feature extraction
• Biometric template is stored in a database or
on an appropriate portable token (e.g. a smart
card)
• There may be several iterations of this process
to refine biometric template

Security of enrolment

• Requirements for enrolment:

 Secure enrolment procedure
 Binding of the biometric template to the
enrolee
 Check of template quality and
matchability
  Fingerprint recognition

• Ridge patterns on fingers uniquely identify

people
• Classification scheme devised in 1890s
• Major features: arch, loop, whorl
• Each fingerprint has at least one of the major
features and many “small features” (so-called
minutiae)

• In an automated system, the sensor must

minimise the image rotation
• Locate minutiae and compare with reference
template
• Minor injuries are a problem
• Liveness detection is important (detached real
fingers, gummy fingers, latent fingerprints)

• Basic steps for fingerprint authentication:

 Image acquisition
 Noise reduction
 Image enhancement
 Feature extraction
 Matching

• Advantages
 Mature technology
 Easy to use/non-intrusive
 High accuracy (comparable to PIN
authentication)
 Long-term stability
 Ability to enrol multiple fingers
 Comparatively low cost
• Disadvantages
 Inability to enrol some users
 Affected by skin condition
 Sensor may get dirty
 Association with forensic applications
  Eye biometrics

• Iris scanning
 Coloured portion of the eye surrounding
the pupil – trabecular meshwork
 Complex iris pattern is used for
authentication
• Retinal scanning
 Retinal vascular pattern on the back inside
the eyeball
 Pattern of blood vessels used for
authentication
• Retinal scanning

• Accurate biometric measure

• Genetic independence: identical twins have
different retinal pattern
• Highly protected, internal organ of the eye
• Advantages
 Potential for high accuracy
 Long-term stability
 Feature is protected from variations
(regarding external environment)
 Genetic independence

• Disadvantages
 Difficult to use
 Intrusive
 Perceived health threat
 High sensor cost

 Iris scanning

• Iris pattern possesses a high degree of

randomness: extremely accurate biometric
• Genetic independence: identical twins have
different iris patterns
• Stable throughout life
• Highly protected, internal organ of the eye
• Patterns can be acquired from a distance (1m)
• Not affected by contact lenses or glasses
• Iris code developed by John Daugman at
Cambridge University
• Extremely low error rates
• Fast processing
• Monitoring of pupil’s oscillation to prevent
fraud
• Monitoring of reflections from the moist
cornea of the living eye

• Advantages
• Potential for high accuracy
• Resistance to impostors
• Long term stability
• Fast processing

• Disadvantages
• Intrusive
• Some people think the state of health can
be detected
• High cost

 Face biometrics

• Static controlled or dynamic uncontrolled

shots
• Visible spectrum or infrared (thermograms)
• Non-invasive, hands-free, and widely accepted
• Questionable discriminatory capability
• Visible spectrum: inexpensive
• Most popular approaches:
• Eigenfaces,
• Local feature analysis.
• Affected by pose, expression, hairstyle, make-
up, lighting, glasses
• Not a reliable biometric measure

• Advantages
 Non-intrusive
 Low cost
 Ability to operate covertly

• Disadvantages
 Affected by appearance and environment
 Low accuracy
 Identical twins attack
 Potential for privacy abuse

 Signature recognition

• Handwritten signatures are an accepted way to

authenticate a person
• Automatic signature recognition measures the
dynamics of the signing process
• Signature generating process is a trained reflex
- imitation difficult especially ‘in real time’

Dynamic signature recognition

 Variety of characteristics can be used:

 angle of the pen,
 pressure of the pen,
 total signing time,
 velocity and acceleration,
 geometry.

• Advantages
 Resistance to forgery
 Widely accepted
 Non-intrusive
 No record of the signature

• Disadvantages
 Signature inconsistencies
 Difficult to use
 Large templates (1K to 3K)
 Problem with trivial signatures
  Speaker verification

• Linguistic and speaker dependent acoustic

patterns
• Speaker’s patterns reflect:
 anatomy (size and shape of mouth and
throat),
 behavioural (voice pitch, speaking style)
• Heavy signal processing involved (spectral
analysis, periodicity, etc.)
• Text-dependent: predetermined set of phrases
for enrolment and identification
• Text-prompted: fixed set of words, but user
prompted to avoid recorded attacks
• Text-independent: free speech, more difficult
to accomplish

• Advantages
 Use of existing telephony infrastructure or
simple microphones
 Easy to use/non-intrusive/hands free
 No negative association

• Disadvantages
 Pre-recorded attack
 Variability of the voice (ill or drunk)
  Affected by background noise
 Large template (5K to 10K)
 Low accuracy

 Biometric Application

• Access control
 To devices
 Cellular phones
 Logging in to computer, laptop, or
PDA
 Cars
 Guns, gun safes
 To local services
  Debitting money from cash dispenser
 Accessing data on smartcard
 To remote services
 E-commerce
 E-business
• Physical access control
 To high security areas
 To public buildings or areas
• Time & attendance control
• Identification
 Forensic person investigation
 Social services applications, e.g.
immigration or prevention of welfare
fraud
 Personal documents, e.g. electronic
drivers license or ID card

 Which biometric method /

product is best?

• Depends on the application

reliability
security
performance
cost
user acceptance
liveness detection
users that are unsuitable
size of sensor