An Overview of Biometrics
Implementasi Access Control
Dr. Mohammad Iqbal
An Overview of
Biometrics
Konten Sistem Biometrik
1.
2.
3.
4.
5.
6.
7.
2
Introduction
Biometric identifiers
Classification of biometrics methods
Biometric system architecture
Performance evaluation
Biometrics System : Signature recognition, Voice
recognition, Retinal scan, Iris scan, FaceFace-scan and
facial thermograph, Hand geometry, DNA
Multimodal Biometrics system
Personal Identification
Association of an individual with an identity:
• Verification (or authentication): confirms or denies a
claimed identity.
• Identification (or recognition): establishes the identity
of a subject (usually from a set of enrolled persons).
3
Obyek Personal Identification
• TokenToken-based:
“something that you have”
• KnowledgeKnowledge-based:
“something that you know”
• BiometricsBiometrics-based:
“something that you are”
4
Defenisi Biometrik
• Bio + metrics:
– The statistical measurement of biological data.
• Biometric Consortium definition:
– Automatically recognizing a person using
distinguishing traits.
5
Domain Aplikasi (I)
•
Access control
– to devices
• cellular phones
• logging into a computer, laptop, or PDA
• cars
• guns
– to local services
• money from a ATM machine
• logging in to computer
• accessing data on smartcard
– to remote services
• e-commerce
• e-business
6
Domain Aplikasi (II)
• 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
– perso
personal
nal documents,
documents, e.g. electronic drivers license
or ID card
7
Biometric Identifiers
Ideal Properties
•
•
•
•
8
Universality
Uniqueness
Stability
Quantitative
Considerations
• Performance
• Acceptability
• Forge resistance
Teknologi Biometrik
• Covered in ISO/IEC 27N2949:
–
–
–
–
–
–
–
9
recognition of signatures,
fingerprint analysis,
speaker recognition,
retinal scan,
iris scan,
face recognition,
hand geometry.
• Found in the
literature:
– vein recognition
(hand),
– keystroke
dynamics,
– palm print,
– gait recognition,
– ear shape.
Klasifikasi Metode Biometrik
Two main classes
Physiological - related to
the shape of the body
– Fingerprints used >100
years
– Palm prints
– Hand geometry
– Hand veins
– Iris recognition
– Retina scan
– Ear canal
– Face recognition
– Facial thermogram
– DNA
• Behavioral - related to
the behavior of a person.
– Signature
– Keystroke dynamics
– Voice
• Static:
–
–
–
–
fingerprint
retinal scan
iris scan
hand geometry
•
Dynamic:
– signature
recognition
– speaker
recognition
Perbandingan Beberapa Teknologi
Biometrik
Human characteristics can be used for biometrics
in terms of the following parameters:
• Universality each person should have the characteristic
• Uniqueness can the biometric separate one individual from another
• Permanence measures how well a biometric resists aging.
• Collectability whether a biometric can be measured quantitatively
• Performance accuracy, speed, and robustness of technology used
• Acceptability degree of approval of a technology
• Circumvention ease of use of a substitute
Perbandingan Beberapa Teknologi
Biometric
Biometrics:
Face
Fingerprint
Hand geometry
Keystrokes
Hand veins
Iris
Retinal scan
Signature
Voice
Facial thermograph
Odor
DNA
Gait
Ear Canal
according to A. K. Jain
(H=High, M=Medium, L=Low)
Univer- Unique- Perman- Collect- Perform- Accept- Circumsality
ness
ence
ability
ance
ability vention*
H
L
M
H
L
H
L
M
H
H
M
H
M
H
M
M
M
H
M
M
M
L
L
L
M
L
M
M
M
M
M
M
M
M
H
H
H
H
M
H
L
H
H
H
M
L
H
L
H
L
L
L
H
L
H
L
M
L
L
M
L
H
L
H
H
L
H
M
H
H
H
H
H
L
L
M
L
H
H
H
L
H
L
L
M
L
L
H
L
H
M
M
M
H
M
M
H
M
Langkah-langkah dalam
LangkahSistem Biometrik
• Enrollment:
Enrollment: The first time you use a biometric
system, it records basic information about you, like
your name or an identification number. It then
captures an image or recording of your specific trait.
• Storage:
Storage: Contrary to what you may see in movies,
most systems don't store the complete image or
recording. They instead analyze your trait and
translate it into a code or graph. Some systems also
record this data onto a smart card that you carry with
you.
• Comparison:
Comparison: The next time you use the system, it
compares the trait you present to the information on
file. Then, it either accepts or rejects that you are who
you claim to be.
Komponen Sistem Biometrik
• A sensor that detects the characteristic being
used for identification
• A computer that reads and stores the
information
• Software that analyzes the characteristic,
translates it into a graph or code and performs
the actual comparisons
Diagram Blok Sistem Biometrik
15
Enrolment
• Capturing, processing and storing of the
biometric template.
• Crucial for the system performance.
• Requirements for enrolment:
– secure enrolment procedure,
– check of template quality and ““matchability
matchability”,
”,
– binding of the biometric template to the person
being enrolled.
16
Diagram Blok Sistem Biometrik
Model Sistem Biometrik
Data
collection
Raw data
Signal
processing
Extracted
features
matching
score
Application
18
Authentication decision
decision
template
storage
Arsitektur Sistem Biometrik
• Basic modules of a biometric system:
1.
2.
3.
4.
5.
19
Data acquisition
Feature extraction
Matching
Decision
Storage
1. Modul Data Acquisition
• Reads the biometric info from the user.
• Examples: video camera, fingerprint
scanner/sensor, microphone, etc.
• All sensors in a given system must be similar
to ensure recognition at any location.
• Environmental conditions may affect their
performance.
20
2. Modul Feature Extraction
• Discriminating features extracted from the
raw biometric data.
• Raw data transformed into small set of bytes –
storage and matching.
• Various ways of extracting the features.
• Pre
Pre--processing of raw data usually necessary.
21
3. Modul Matching
• The core of the biometric system.
• Measures the similarity of the claimant’s
sample with a reference template.
• Typical methods: distance metrics,
probabilistic measures, neural networks, etc.
• The result: a number known as match score.
22
4. Modul Decision
• Interprets the match score from the matching
module.
• 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.
23
5. Storage Module
• Maintains the templates for enrolled users.
• One or more templates for each user.
• The templates may be stored in:
– a special component in the biometric device,
– conventional computer database,
– portable memories such as smartcards.
24
Pengukuran Kinerja
•false accept rate (FAR) or false match rate (FMR):
(FMR): the
probability that the system incorrectly declares a
successful match between the input pattern and a nonnonmatching pattern in the database. Measures the
percent of invalid inputs being accepted.
•false reject rate (FRR) or false nonnon-match rate
(FNMR):: the probability that the system incorrectly
(FNMR)
declares failure of match between the input pattern
and the matching template in the database. Measures
the percent of valid inputs being rejected.
•relative operating characteristic (ROC):
(ROC): In general,
the matching algorithm performs a decision using a
threshold. In biometric systems the FAR and FRR can
typically be traded off against each other by changing
those parameters.
Pengukuran Kinerja
•equal error rate (EER):
(EER): the rate at which both accept
and reject errors are equal. The lower the EER, the
more accurate the system is considered to be.
•failure to enroll rate (FTE or FER):
FER): the percentage of
data input is considered invalid and fails to input into
the system. Failure to enroll happens when the data
obtained by the sensor are considered invalid or of
poor quality.
•failure to capture rate (FTC):
(FTC): Within automatic
systems, the probability that the system fails to detect
a biometric characteristic when presented correctly.
•template capacity:
capacity: the maximum number of sets of
data which can be input in to the system.
Kemungkinan Keputusan dari
Sistem Biometrik
•
•
•
•
27
A genuine individual is accepted.
A genuine individual is rejected (error).
An impostor is rejected.
An impostor is accepted (error).
Errors
• Balance needed between 2 types of error:
– Type I:
I: system fails to recognize valid user (‘false
non--match’ or ‘false rejection’).
non
– Type II:
II: system accepts impostor (‘false match’ or
‘false acceptance’).
• Application dependent trade
trade--off between two
error types.
28
Tolerance Threshold
• Error tolerance threshold is crucial and
application dependent.
• Tolerance too large gives Type II error (admit
impostors).
• Tolerance too small gives Type I errors (reject
legitimate users).
• Equal error rate for comparison: false nonnonmatch equal to false match.
29
State of the Art of
Biometric Recognition Systems
Biometrics
Equal False False
Error Accept Reject Subjects
Ratio Ratio Ratio
Face
n.a.
1%
10%
37437
Fingerprint
n.a.
1%
0.1%
25000
Fingerprint
2%
2%
2%
100
Hand geometry
1%
2%
0.1%
129
Iris
< 1%
0.94% 0.99%
1224
Iris
0.01% 0.0001% 0.2%
132
Keystrokes
1.8%
7%
0.1%
15
6%
2%
10%
310
Voice
Comment
Varied lighting,
indoor/outdoor
US Government
operational data
Rotation and exaggerated
skin distortion
With rings and improper
placement
Indoor environment
Best conditions
During 6 months period
Text independent,
multilingual
Reference
FRVT (2002)
FpVTE (2003)
FVC (2004)
(2005)
ITIRT (2005)
NIST (2005)
(2005)
NIST (2004)
Biometric Technologies
•
•
•
•
•
•
•
31
Signature recognition
Voice recognition
Retinal scan
Iris scan
Face biometrics
Hand geometry
DNA
1. Signature Recognition
• Signatures in wide use for many years.
• Signature generating process a trained reflex imitation difficult especially ‘in real time’.
• Automatic signature recognition measures the
dynamics of the signing process.
32
Dynamic Signature Recognition
• Variety of characteristics can be used:
–
–
–
–
–
33
angle of the pen,
pressure of the pen,
total signing time,
velocity and acceleration,
geometry.
Dynamic Signature Verification
(I)
Electronic pen [LCI-SmartPen]
34
Dynamic Signature Verification
(II)
Digitising tablet by
Wacom Technologies
35
Digitising tablet [Hesy Signature Pad
by BS Biometric Systems GmbH]
Signature Recognition:
Advantages / Disadvantages
• Advantages:
–
–
–
–
Resistance to forgery
Widely accepted
Non
Non--intrusive
No record of the signature
• Disadvantages:
– Signature inconsistencies
– Difficult to use
– Large templates (1K to 3K)
36
2. 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.
37
Features of Fingerprints
Fingerprint Information
Features of Fingerprints
39
Fingerprint Recognition (cont.)
• In a machine system, reader must minimize
image rotation.
• Look for minutiae and compare.
• Minor injuries a problem.
• Automatic systems can not be defrauded by
detached real fingers.
40
Fingerprint Authentication
• Basic steps for fingerprint authentication:
–
–
–
–
–
41
Image acquisition,
Noise reduction,
Image enhancement,
Feature extraction,
Matching.
Fingerprint Processing
a
b
a) Original
c
d
b) Orientation
c) Binarised
d) Thinned
e
f
e) Minutiae
f) Minutia graph
42
Fingerprint Recognition
• Sensors
–
–
–
–
optical sensors
ultrasound sensors
chip
chip--based sensors
thermal sensors
• Integrated products
– for identification – AFIS systems
– for verification
43
Fingerprint Recognition:
Sensors
Optical fingerprint sensor
[Fingerprint Identification Unit
FIU-001/500 by Sony]
Electro-optical sensor
[DELSY® CMOS sensor modul]
Capacitive sensor
[FingerTIP™ by Infineon]
44
Fingerprint Recognition:
Sensors
Optical technology
Light
source
Finger
Prism
Lens
Video Camera (CCD)
Light reflects from the surface of the prism where the finger is not
in contact with it, while it penetrates the surface of the prism
where the finger touches the surface of the prism. The resulting
image goes through a lens into a video camera.
Fingerprint Recognition:
Sensors
Capacity technology
Fingerprint Recognition:
Sensors
Fiber optic technology
Fingerprint Recognition:
Sensors
Thermal sensor
[FingerChip™ by ATMEL
(was: Thomson CSF)]
E-Field Sensor
[FingerLoc™ by Authentec]
48
Fingerprint Recognition:
Integrated Systems
[BioMouse™ Plus by American Biometric Company]
Physical Access Control System
[BioGate Tower by Bergdata]
49
[ID Mouse by Siemens]
Fingerprint Recognition:
Integrated Systems
Keyboard [G 81-12000
by Cherry]
[TravelMate 740 by Compaq und Acer]
50
System including
fingerprint sensor,
smartcard reader and
display by DELSY
Fingerprint Recognition:
Advantages / Disadvantages
• Advantages:
–
–
–
–
–
Mature technology
Easy to use/non
use/non--intrusive
High accuracy
Long
Long--term stability
Ability to enrol multiple fingers
• Disadvantages:
– Inability to enrol some users
– Affected by skin condition
– Association with forensic applications
51
3. Speech Recognition
• Linguistic and speaker dependent acoustic
patterns.
• Speaker’s patterns reflect:
– anatomy (size and shape of mouth and throat),
– behavioral (voice pitch, speaking style).
• Heavy signal processing involved (spectral
analysis, periodicity, etc)
52
Speaker Recognition Systems
• Text
Text--dependent: predetermined set of phrases
for enrolment and identification.
• Text
Text--prompted: fixed set of words, but user
prompted to avoid recorded attacks.
• Text
Text--independent: free speech, more difficult
to accomplish.
53
Voice Recognition Credit Card
• The card would have
a button on it and
when pressed it
would say "Please
say your password".
• Compares voice to
data file store
remotely (via the
internet)
Uses of Voice Recognition
Technology
• Access Control
• Computer work
stations
• Time Clocks
Speaker Recognition:
Advantages/ Disadvantages
• Advantages:
– Use of existing telephony infrastructure
– Easy to use/non
use/non--intrusive/hands free
– No negative association
• Disadvantages:
–
–
–
–
56
Pre
Pre--recorded attack
Variability of the voice
Affected by noise
Large template (5K to 10K)
4. Eye Biometric
• Retina:
– back inside of the eye ball.
– pattern of blood vessels used for identification
• Iris:
– colored portion of the eye surrounding the pupil.
– complex iris pattern used for identification.
identification.
57
Retinal Pattern
• Accurate biometric
measure.
• Genetically independent:
identical twins have
different retinal pattern.
• Highly protected, internal
organ of the eye.
• May change during the
life of a person.
58
Retinal Recognition
Retinal recognition system [Icam 2001 by Eyedentify]
59
Retinal Challenges
2nd Refresher Course in
Mathematical Science,9th-
Retinal Scan:
Advantages / Disadvantages
• Advantages:
– High accuracy
– Long
Long--term stability
– Fast verification
• Disadvantages:
– Difficult to use
– Intrusive
– Limited applications
61
Iris Properties
• Iris pattern possesses a high degree of randomness:
extremely accurate biometric.
• Genetically independent: identical twins have
different iris pattern.
• Stable throughout life.
• Highly protected, internal organ of the eye.
• Patterns can be acquired from a distance (1m).
• Patterns can be encoded into 256 bytes.
62
Iris Recognition
•
•
•
•
•
63
Iris code developed by John Daugman at Cambridge.
Extremely low error rates.
Fast processing.
Monitoring of pupils oscillation to prevent fraud.
Monitoring of reflections from the moist cornea of
the living eye.
The Iris Code
64
Iris Recognition
System for passive iris recognition by Sensar
System for active iris recognition by IrisScan
65
Iris Challenges
2nd Refresher Course in
Mathematical Science,9th-
Iris Recognition:
Advantages / Disadvantages
• Advantages:
–
–
–
–
High accuracy
Long term stability
Nearly non
non--intrusive
Fast processing
• Disadvantages:
– Not exactly easy to use
– High false non
non--match rates
– High cost
67
5. FaceFace-scan and Facial
Thermographs
• Static controlled or dynamic uncontrolled
shots.
• Visible spectrum or infrared (thermographs).
• NonNon-invasive, handshands-free, and widely
accepted.
• Questionable discriminatory capability.
68
Face Recognition
• Visible spectrum:
inexpensive.
• Most popular approaches:
– eigen faces,
– Local feature analysis.
69
• Affected by pose,
expression, hairstyle,
make--up, lighting,
make
eyeglasses.
• Not a reliable biometric
measure.
Facial Recognition Technology
Face Recognition
Face recognition system
[TrueFace Engine by Miros]
Face recognition system
[One-to-One™ by Biometric Access Corporation]
71
Face Recognition Challenges
72
Face Recognition:
Advantages / Disadvantages
• Advantages:
– Non
Non--intrusive
– Low cost
– Ability to operate covertly
• Disadvantages:
–
–
–
–
73
Affected by appearance/environment
High false non
non--match rates
Identical twins attack
Potential for privacy abuse
Facial Thermograph
• Captures the heat emission patterns derived
from the blood vessels under the skin.
• Infrared camera: unaffected by external
changes (even plastic surgery!) or lighting.
• Unique but accuracy questionable.
• Affected by emotional and health state.
74
Facial Thermograph:
Advantages / Disadvantages
• Advantages:
–
–
–
–
–
Non
Non--intrusive
Stable
Not affected by external changes
Identical twins resistant
Ability to operate covertly
• Disadvantages:
– High cost (infrared camera)
– New technology
– Potential for privacy abuse
75
6. Hand Geometry
• Features: dimensions and shape of the hand,
fingers, and knuckles as well as their relative
locations.
• Two images taken: one from the top and one
from the side.
•Captured using a CCD camera, or LED
•Orthographic Scanning
•Recognition System’s Crossover = 0.1%
76
Hand Geometry Reading
Hand geometry reader by Recognition Systems
Hand geometry reader for
two finger recognition by BioMet Partners
77
Hand Geometry:
Advantages / Disadvantages
• Advantages:
–
–
–
–
Not affected by environment
Mature technology
Non
Non--intrusive
Relatively stable
–
–
–
–
Low accuracy
High cost
Relatively large readers
Difficult to use for some users ((arthritis,
arthritis, missing
fingers or large hands)
• Disadvantages:
78
Vein Geometry
• A person's veins are completely unique.
• Twins don't have identical veins, and a person's veins
differ between their left and right sides.
• Many veins are not visible through the skin, making them
extremely difficult to counterfeit or tamper with. Their
shape also changes very little as a person ages.
Vein scanners use nearnear-infrared light to reveal the patterns
in a person’s veins.
• Place your finger, wrist, palm or the back of your hand on
or near the scanner. A camera takes a digital picture using
near--IR light. The hemoglobin in your blood absorbs the
near
light, so veins appear black in the picture. Software
creates a reference template based on the shape and
location of the vein structure.
• Scanners that analyze vein geometry
are completely different from vein
scanning tests that happen in hospitals.
Vein scans for medical purposes
usually use radioactive particles.
Biometric security scans use light that
is similar to the light that comes from a
remote control.
Image from HowStuffWorks.com and Fujitsu
7. DNA
• The key to DNA evidence lies in comparing the DNA
from the scene of a crime with a suspect's DNA. To do
this, investigators have to do three things:
– Collect DNA from the subject
(intrusive and messy)
– Analyze the DNA to create a DNA profile (slow and costly)
– Compare the profile to a database
(not well populated)
• DNA can be extracted from almost any tissue, including
hair, fingernails, bones, teeth and bodily fluids.
• The most commonly used database in the United States is
mbined DNA Index
called CODIS
CODIS,, which stands for Co
Combined
System. CODIS is maintained by the FBI.
DNA
Strengths
◦ Ultimate unique code for one’s
individuality
◦ Extremely accurate
◦ Used widely for paternity
identification
Weakness
◦ Very Intrusive [disturbing]
◦ Requires cooperative subject
◦ Easily contaminated
◦ Process is too slow for automatic
real time identification
◦ Identical for twins
2nd Refresher Course in
Mathematical Science,9th-
Other methods of Biometrics
Physiological
• Ear shape recognition
• Body odor recognition
• Dental pattern recognition
Behavioral
• Voice print recognition
• Signature recognition
• Keystroke analysis (typing pattern)
Other methods of Biometrics
Keystroke scan:
Measures the time
between strokes
and duration of
key pressed.
–
83
Most commonly
used in systems
where keyboard is
already being
used.
Multimodal Biometric Systems
• Combination of biometric technologies
– Fingerprint and face recognition
– Face recognition and lip movement
– Fingerprint recognition and dynamic signature
verification
• Increase the level of security achieved by the
system
• Enlarge the user base
84
Multimodal Biometrics
Pattern Recognition Concept
Sensors
Biometrics
Extractors
Image- and
signal- pro.
algo.
Classifiers
Data Rep. Feature
1D (wav),
Vectors
Voice, signature
acoustics, face,
2D (bmp,
fingerprint, iris,
tiff, png)
hand geometry, etc
Enrolment
Negotiator
Threshold
Scores
Training
Submission
Decision:
Match,
Non-match,
Inconclusive
An Example
Example::
A MultiMulti-model System
Sensors
Extractors
Classifiers
Negotiator
Accept/
Reject
ID
Face
Extractor
Face
Feature
Face
MLP
AND
2D (bmp)
Voice
Extractor
Voice
Feature
Voice
MLP
1D (wav)
Objective: to build a hybrid and expandable biometric app. prototype
Potential: be a middleware and a research tool
Selesai
Dr. Mohammad Iqbal
An Overview of
Biometrics
Konten Sistem Biometrik
1.
2.
3.
4.
5.
6.
7.
2
Introduction
Biometric identifiers
Classification of biometrics methods
Biometric system architecture
Performance evaluation
Biometrics System : Signature recognition, Voice
recognition, Retinal scan, Iris scan, FaceFace-scan and
facial thermograph, Hand geometry, DNA
Multimodal Biometrics system
Personal Identification
Association of an individual with an identity:
• Verification (or authentication): confirms or denies a
claimed identity.
• Identification (or recognition): establishes the identity
of a subject (usually from a set of enrolled persons).
3
Obyek Personal Identification
• TokenToken-based:
“something that you have”
• KnowledgeKnowledge-based:
“something that you know”
• BiometricsBiometrics-based:
“something that you are”
4
Defenisi Biometrik
• Bio + metrics:
– The statistical measurement of biological data.
• Biometric Consortium definition:
– Automatically recognizing a person using
distinguishing traits.
5
Domain Aplikasi (I)
•
Access control
– to devices
• cellular phones
• logging into a computer, laptop, or PDA
• cars
• guns
– to local services
• money from a ATM machine
• logging in to computer
• accessing data on smartcard
– to remote services
• e-commerce
• e-business
6
Domain Aplikasi (II)
• 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
– perso
personal
nal documents,
documents, e.g. electronic drivers license
or ID card
7
Biometric Identifiers
Ideal Properties
•
•
•
•
8
Universality
Uniqueness
Stability
Quantitative
Considerations
• Performance
• Acceptability
• Forge resistance
Teknologi Biometrik
• Covered in ISO/IEC 27N2949:
–
–
–
–
–
–
–
9
recognition of signatures,
fingerprint analysis,
speaker recognition,
retinal scan,
iris scan,
face recognition,
hand geometry.
• Found in the
literature:
– vein recognition
(hand),
– keystroke
dynamics,
– palm print,
– gait recognition,
– ear shape.
Klasifikasi Metode Biometrik
Two main classes
Physiological - related to
the shape of the body
– Fingerprints used >100
years
– Palm prints
– Hand geometry
– Hand veins
– Iris recognition
– Retina scan
– Ear canal
– Face recognition
– Facial thermogram
– DNA
• Behavioral - related to
the behavior of a person.
– Signature
– Keystroke dynamics
– Voice
• Static:
–
–
–
–
fingerprint
retinal scan
iris scan
hand geometry
•
Dynamic:
– signature
recognition
– speaker
recognition
Perbandingan Beberapa Teknologi
Biometrik
Human characteristics can be used for biometrics
in terms of the following parameters:
• Universality each person should have the characteristic
• Uniqueness can the biometric separate one individual from another
• Permanence measures how well a biometric resists aging.
• Collectability whether a biometric can be measured quantitatively
• Performance accuracy, speed, and robustness of technology used
• Acceptability degree of approval of a technology
• Circumvention ease of use of a substitute
Perbandingan Beberapa Teknologi
Biometric
Biometrics:
Face
Fingerprint
Hand geometry
Keystrokes
Hand veins
Iris
Retinal scan
Signature
Voice
Facial thermograph
Odor
DNA
Gait
Ear Canal
according to A. K. Jain
(H=High, M=Medium, L=Low)
Univer- Unique- Perman- Collect- Perform- Accept- Circumsality
ness
ence
ability
ance
ability vention*
H
L
M
H
L
H
L
M
H
H
M
H
M
H
M
M
M
H
M
M
M
L
L
L
M
L
M
M
M
M
M
M
M
M
H
H
H
H
M
H
L
H
H
H
M
L
H
L
H
L
L
L
H
L
H
L
M
L
L
M
L
H
L
H
H
L
H
M
H
H
H
H
H
L
L
M
L
H
H
H
L
H
L
L
M
L
L
H
L
H
M
M
M
H
M
M
H
M
Langkah-langkah dalam
LangkahSistem Biometrik
• Enrollment:
Enrollment: The first time you use a biometric
system, it records basic information about you, like
your name or an identification number. It then
captures an image or recording of your specific trait.
• Storage:
Storage: Contrary to what you may see in movies,
most systems don't store the complete image or
recording. They instead analyze your trait and
translate it into a code or graph. Some systems also
record this data onto a smart card that you carry with
you.
• Comparison:
Comparison: The next time you use the system, it
compares the trait you present to the information on
file. Then, it either accepts or rejects that you are who
you claim to be.
Komponen Sistem Biometrik
• A sensor that detects the characteristic being
used for identification
• A computer that reads and stores the
information
• Software that analyzes the characteristic,
translates it into a graph or code and performs
the actual comparisons
Diagram Blok Sistem Biometrik
15
Enrolment
• Capturing, processing and storing of the
biometric template.
• Crucial for the system performance.
• Requirements for enrolment:
– secure enrolment procedure,
– check of template quality and ““matchability
matchability”,
”,
– binding of the biometric template to the person
being enrolled.
16
Diagram Blok Sistem Biometrik
Model Sistem Biometrik
Data
collection
Raw data
Signal
processing
Extracted
features
matching
score
Application
18
Authentication decision
decision
template
storage
Arsitektur Sistem Biometrik
• Basic modules of a biometric system:
1.
2.
3.
4.
5.
19
Data acquisition
Feature extraction
Matching
Decision
Storage
1. Modul Data Acquisition
• Reads the biometric info from the user.
• Examples: video camera, fingerprint
scanner/sensor, microphone, etc.
• All sensors in a given system must be similar
to ensure recognition at any location.
• Environmental conditions may affect their
performance.
20
2. Modul Feature Extraction
• Discriminating features extracted from the
raw biometric data.
• Raw data transformed into small set of bytes –
storage and matching.
• Various ways of extracting the features.
• Pre
Pre--processing of raw data usually necessary.
21
3. Modul Matching
• The core of the biometric system.
• Measures the similarity of the claimant’s
sample with a reference template.
• Typical methods: distance metrics,
probabilistic measures, neural networks, etc.
• The result: a number known as match score.
22
4. Modul Decision
• Interprets the match score from the matching
module.
• 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.
23
5. Storage Module
• Maintains the templates for enrolled users.
• One or more templates for each user.
• The templates may be stored in:
– a special component in the biometric device,
– conventional computer database,
– portable memories such as smartcards.
24
Pengukuran Kinerja
•false accept rate (FAR) or false match rate (FMR):
(FMR): the
probability that the system incorrectly declares a
successful match between the input pattern and a nonnonmatching pattern in the database. Measures the
percent of invalid inputs being accepted.
•false reject rate (FRR) or false nonnon-match rate
(FNMR):: the probability that the system incorrectly
(FNMR)
declares failure of match between the input pattern
and the matching template in the database. Measures
the percent of valid inputs being rejected.
•relative operating characteristic (ROC):
(ROC): In general,
the matching algorithm performs a decision using a
threshold. In biometric systems the FAR and FRR can
typically be traded off against each other by changing
those parameters.
Pengukuran Kinerja
•equal error rate (EER):
(EER): the rate at which both accept
and reject errors are equal. The lower the EER, the
more accurate the system is considered to be.
•failure to enroll rate (FTE or FER):
FER): the percentage of
data input is considered invalid and fails to input into
the system. Failure to enroll happens when the data
obtained by the sensor are considered invalid or of
poor quality.
•failure to capture rate (FTC):
(FTC): Within automatic
systems, the probability that the system fails to detect
a biometric characteristic when presented correctly.
•template capacity:
capacity: the maximum number of sets of
data which can be input in to the system.
Kemungkinan Keputusan dari
Sistem Biometrik
•
•
•
•
27
A genuine individual is accepted.
A genuine individual is rejected (error).
An impostor is rejected.
An impostor is accepted (error).
Errors
• Balance needed between 2 types of error:
– Type I:
I: system fails to recognize valid user (‘false
non--match’ or ‘false rejection’).
non
– Type II:
II: system accepts impostor (‘false match’ or
‘false acceptance’).
• Application dependent trade
trade--off between two
error types.
28
Tolerance Threshold
• Error tolerance threshold is crucial and
application dependent.
• Tolerance too large gives Type II error (admit
impostors).
• Tolerance too small gives Type I errors (reject
legitimate users).
• Equal error rate for comparison: false nonnonmatch equal to false match.
29
State of the Art of
Biometric Recognition Systems
Biometrics
Equal False False
Error Accept Reject Subjects
Ratio Ratio Ratio
Face
n.a.
1%
10%
37437
Fingerprint
n.a.
1%
0.1%
25000
Fingerprint
2%
2%
2%
100
Hand geometry
1%
2%
0.1%
129
Iris
< 1%
0.94% 0.99%
1224
Iris
0.01% 0.0001% 0.2%
132
Keystrokes
1.8%
7%
0.1%
15
6%
2%
10%
310
Voice
Comment
Varied lighting,
indoor/outdoor
US Government
operational data
Rotation and exaggerated
skin distortion
With rings and improper
placement
Indoor environment
Best conditions
During 6 months period
Text independent,
multilingual
Reference
FRVT (2002)
FpVTE (2003)
FVC (2004)
(2005)
ITIRT (2005)
NIST (2005)
(2005)
NIST (2004)
Biometric Technologies
•
•
•
•
•
•
•
31
Signature recognition
Voice recognition
Retinal scan
Iris scan
Face biometrics
Hand geometry
DNA
1. Signature Recognition
• Signatures in wide use for many years.
• Signature generating process a trained reflex imitation difficult especially ‘in real time’.
• Automatic signature recognition measures the
dynamics of the signing process.
32
Dynamic Signature Recognition
• Variety of characteristics can be used:
–
–
–
–
–
33
angle of the pen,
pressure of the pen,
total signing time,
velocity and acceleration,
geometry.
Dynamic Signature Verification
(I)
Electronic pen [LCI-SmartPen]
34
Dynamic Signature Verification
(II)
Digitising tablet by
Wacom Technologies
35
Digitising tablet [Hesy Signature Pad
by BS Biometric Systems GmbH]
Signature Recognition:
Advantages / Disadvantages
• Advantages:
–
–
–
–
Resistance to forgery
Widely accepted
Non
Non--intrusive
No record of the signature
• Disadvantages:
– Signature inconsistencies
– Difficult to use
– Large templates (1K to 3K)
36
2. 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.
37
Features of Fingerprints
Fingerprint Information
Features of Fingerprints
39
Fingerprint Recognition (cont.)
• In a machine system, reader must minimize
image rotation.
• Look for minutiae and compare.
• Minor injuries a problem.
• Automatic systems can not be defrauded by
detached real fingers.
40
Fingerprint Authentication
• Basic steps for fingerprint authentication:
–
–
–
–
–
41
Image acquisition,
Noise reduction,
Image enhancement,
Feature extraction,
Matching.
Fingerprint Processing
a
b
a) Original
c
d
b) Orientation
c) Binarised
d) Thinned
e
f
e) Minutiae
f) Minutia graph
42
Fingerprint Recognition
• Sensors
–
–
–
–
optical sensors
ultrasound sensors
chip
chip--based sensors
thermal sensors
• Integrated products
– for identification – AFIS systems
– for verification
43
Fingerprint Recognition:
Sensors
Optical fingerprint sensor
[Fingerprint Identification Unit
FIU-001/500 by Sony]
Electro-optical sensor
[DELSY® CMOS sensor modul]
Capacitive sensor
[FingerTIP™ by Infineon]
44
Fingerprint Recognition:
Sensors
Optical technology
Light
source
Finger
Prism
Lens
Video Camera (CCD)
Light reflects from the surface of the prism where the finger is not
in contact with it, while it penetrates the surface of the prism
where the finger touches the surface of the prism. The resulting
image goes through a lens into a video camera.
Fingerprint Recognition:
Sensors
Capacity technology
Fingerprint Recognition:
Sensors
Fiber optic technology
Fingerprint Recognition:
Sensors
Thermal sensor
[FingerChip™ by ATMEL
(was: Thomson CSF)]
E-Field Sensor
[FingerLoc™ by Authentec]
48
Fingerprint Recognition:
Integrated Systems
[BioMouse™ Plus by American Biometric Company]
Physical Access Control System
[BioGate Tower by Bergdata]
49
[ID Mouse by Siemens]
Fingerprint Recognition:
Integrated Systems
Keyboard [G 81-12000
by Cherry]
[TravelMate 740 by Compaq und Acer]
50
System including
fingerprint sensor,
smartcard reader and
display by DELSY
Fingerprint Recognition:
Advantages / Disadvantages
• Advantages:
–
–
–
–
–
Mature technology
Easy to use/non
use/non--intrusive
High accuracy
Long
Long--term stability
Ability to enrol multiple fingers
• Disadvantages:
– Inability to enrol some users
– Affected by skin condition
– Association with forensic applications
51
3. Speech Recognition
• Linguistic and speaker dependent acoustic
patterns.
• Speaker’s patterns reflect:
– anatomy (size and shape of mouth and throat),
– behavioral (voice pitch, speaking style).
• Heavy signal processing involved (spectral
analysis, periodicity, etc)
52
Speaker Recognition Systems
• Text
Text--dependent: predetermined set of phrases
for enrolment and identification.
• Text
Text--prompted: fixed set of words, but user
prompted to avoid recorded attacks.
• Text
Text--independent: free speech, more difficult
to accomplish.
53
Voice Recognition Credit Card
• The card would have
a button on it and
when pressed it
would say "Please
say your password".
• Compares voice to
data file store
remotely (via the
internet)
Uses of Voice Recognition
Technology
• Access Control
• Computer work
stations
• Time Clocks
Speaker Recognition:
Advantages/ Disadvantages
• Advantages:
– Use of existing telephony infrastructure
– Easy to use/non
use/non--intrusive/hands free
– No negative association
• Disadvantages:
–
–
–
–
56
Pre
Pre--recorded attack
Variability of the voice
Affected by noise
Large template (5K to 10K)
4. Eye Biometric
• Retina:
– back inside of the eye ball.
– pattern of blood vessels used for identification
• Iris:
– colored portion of the eye surrounding the pupil.
– complex iris pattern used for identification.
identification.
57
Retinal Pattern
• Accurate biometric
measure.
• Genetically independent:
identical twins have
different retinal pattern.
• Highly protected, internal
organ of the eye.
• May change during the
life of a person.
58
Retinal Recognition
Retinal recognition system [Icam 2001 by Eyedentify]
59
Retinal Challenges
2nd Refresher Course in
Mathematical Science,9th-
Retinal Scan:
Advantages / Disadvantages
• Advantages:
– High accuracy
– Long
Long--term stability
– Fast verification
• Disadvantages:
– Difficult to use
– Intrusive
– Limited applications
61
Iris Properties
• Iris pattern possesses a high degree of randomness:
extremely accurate biometric.
• Genetically independent: identical twins have
different iris pattern.
• Stable throughout life.
• Highly protected, internal organ of the eye.
• Patterns can be acquired from a distance (1m).
• Patterns can be encoded into 256 bytes.
62
Iris Recognition
•
•
•
•
•
63
Iris code developed by John Daugman at Cambridge.
Extremely low error rates.
Fast processing.
Monitoring of pupils oscillation to prevent fraud.
Monitoring of reflections from the moist cornea of
the living eye.
The Iris Code
64
Iris Recognition
System for passive iris recognition by Sensar
System for active iris recognition by IrisScan
65
Iris Challenges
2nd Refresher Course in
Mathematical Science,9th-
Iris Recognition:
Advantages / Disadvantages
• Advantages:
–
–
–
–
High accuracy
Long term stability
Nearly non
non--intrusive
Fast processing
• Disadvantages:
– Not exactly easy to use
– High false non
non--match rates
– High cost
67
5. FaceFace-scan and Facial
Thermographs
• Static controlled or dynamic uncontrolled
shots.
• Visible spectrum or infrared (thermographs).
• NonNon-invasive, handshands-free, and widely
accepted.
• Questionable discriminatory capability.
68
Face Recognition
• Visible spectrum:
inexpensive.
• Most popular approaches:
– eigen faces,
– Local feature analysis.
69
• Affected by pose,
expression, hairstyle,
make--up, lighting,
make
eyeglasses.
• Not a reliable biometric
measure.
Facial Recognition Technology
Face Recognition
Face recognition system
[TrueFace Engine by Miros]
Face recognition system
[One-to-One™ by Biometric Access Corporation]
71
Face Recognition Challenges
72
Face Recognition:
Advantages / Disadvantages
• Advantages:
– Non
Non--intrusive
– Low cost
– Ability to operate covertly
• Disadvantages:
–
–
–
–
73
Affected by appearance/environment
High false non
non--match rates
Identical twins attack
Potential for privacy abuse
Facial Thermograph
• Captures the heat emission patterns derived
from the blood vessels under the skin.
• Infrared camera: unaffected by external
changes (even plastic surgery!) or lighting.
• Unique but accuracy questionable.
• Affected by emotional and health state.
74
Facial Thermograph:
Advantages / Disadvantages
• Advantages:
–
–
–
–
–
Non
Non--intrusive
Stable
Not affected by external changes
Identical twins resistant
Ability to operate covertly
• Disadvantages:
– High cost (infrared camera)
– New technology
– Potential for privacy abuse
75
6. Hand Geometry
• Features: dimensions and shape of the hand,
fingers, and knuckles as well as their relative
locations.
• Two images taken: one from the top and one
from the side.
•Captured using a CCD camera, or LED
•Orthographic Scanning
•Recognition System’s Crossover = 0.1%
76
Hand Geometry Reading
Hand geometry reader by Recognition Systems
Hand geometry reader for
two finger recognition by BioMet Partners
77
Hand Geometry:
Advantages / Disadvantages
• Advantages:
–
–
–
–
Not affected by environment
Mature technology
Non
Non--intrusive
Relatively stable
–
–
–
–
Low accuracy
High cost
Relatively large readers
Difficult to use for some users ((arthritis,
arthritis, missing
fingers or large hands)
• Disadvantages:
78
Vein Geometry
• A person's veins are completely unique.
• Twins don't have identical veins, and a person's veins
differ between their left and right sides.
• Many veins are not visible through the skin, making them
extremely difficult to counterfeit or tamper with. Their
shape also changes very little as a person ages.
Vein scanners use nearnear-infrared light to reveal the patterns
in a person’s veins.
• Place your finger, wrist, palm or the back of your hand on
or near the scanner. A camera takes a digital picture using
near--IR light. The hemoglobin in your blood absorbs the
near
light, so veins appear black in the picture. Software
creates a reference template based on the shape and
location of the vein structure.
• Scanners that analyze vein geometry
are completely different from vein
scanning tests that happen in hospitals.
Vein scans for medical purposes
usually use radioactive particles.
Biometric security scans use light that
is similar to the light that comes from a
remote control.
Image from HowStuffWorks.com and Fujitsu
7. DNA
• The key to DNA evidence lies in comparing the DNA
from the scene of a crime with a suspect's DNA. To do
this, investigators have to do three things:
– Collect DNA from the subject
(intrusive and messy)
– Analyze the DNA to create a DNA profile (slow and costly)
– Compare the profile to a database
(not well populated)
• DNA can be extracted from almost any tissue, including
hair, fingernails, bones, teeth and bodily fluids.
• The most commonly used database in the United States is
mbined DNA Index
called CODIS
CODIS,, which stands for Co
Combined
System. CODIS is maintained by the FBI.
DNA
Strengths
◦ Ultimate unique code for one’s
individuality
◦ Extremely accurate
◦ Used widely for paternity
identification
Weakness
◦ Very Intrusive [disturbing]
◦ Requires cooperative subject
◦ Easily contaminated
◦ Process is too slow for automatic
real time identification
◦ Identical for twins
2nd Refresher Course in
Mathematical Science,9th-
Other methods of Biometrics
Physiological
• Ear shape recognition
• Body odor recognition
• Dental pattern recognition
Behavioral
• Voice print recognition
• Signature recognition
• Keystroke analysis (typing pattern)
Other methods of Biometrics
Keystroke scan:
Measures the time
between strokes
and duration of
key pressed.
–
83
Most commonly
used in systems
where keyboard is
already being
used.
Multimodal Biometric Systems
• Combination of biometric technologies
– Fingerprint and face recognition
– Face recognition and lip movement
– Fingerprint recognition and dynamic signature
verification
• Increase the level of security achieved by the
system
• Enlarge the user base
84
Multimodal Biometrics
Pattern Recognition Concept
Sensors
Biometrics
Extractors
Image- and
signal- pro.
algo.
Classifiers
Data Rep. Feature
1D (wav),
Vectors
Voice, signature
acoustics, face,
2D (bmp,
fingerprint, iris,
tiff, png)
hand geometry, etc
Enrolment
Negotiator
Threshold
Scores
Training
Submission
Decision:
Match,
Non-match,
Inconclusive
An Example
Example::
A MultiMulti-model System
Sensors
Extractors
Classifiers
Negotiator
Accept/
Reject
ID
Face
Extractor
Face
Feature
Face
MLP
AND
2D (bmp)
Voice
Extractor
Voice
Feature
Voice
MLP
1D (wav)
Objective: to build a hybrid and expandable biometric app. prototype
Potential: be a middleware and a research tool
Selesai