Date post: | 09-Apr-2018 |
Category: |
Documents |
Upload: | sruthi-reddy-chada |
View: | 220 times |
Download: | 0 times |
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 1/17
BIOMETRICS
KAMALA INSTITUTE OF TECHNOLOGY & SCIENCE
SINGAPUR, HUZURABAD
2008-2009
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 2/17
ABSTRACT
Biometrics refers to the automatic
identification of a person based on his/her
physiological or behavioral characteristics.
This method of identification is preferred
over traditional methods involving
password’s and PIN numbers.
Using Biometrics you can unlock
your houses, withdrawing money from a
bank with just a blink of an eye, a tap of
your finger or by just showing your face.
Fingerprint recognition is most widely
accepted biometric among the technology
being used today. Biometrics in Face
recognition has received a surge of
attention since of disaster of 11/9 for its
ability to identify known terrorists and
criminals.
First usage:-
First used in China in the 14th
century -- merchants stamped hand and
foot prints of children on paper with ink to
distinguish one child from another. In
Europe, Richard Edward Henry developed
fingerprinting for Scotland Yard in the
later half of the 19th century.
INTRODUCTION
Definition: Biometrics is the process by
which distinguishing human anatomy is
used for identification and verification.
Biometric systems are pattern recognition
systems which are automated methods of
determining the authenticity of a specific
physiological or behavioral characteristic
possessed by the user to determine or
verify identify. The word biometrics
comes from the Greek words bio -
meaning life, and metric - meaning to
measure.
Identification is determining who a
person is. One has to establish a person's
identity (Who am I?).
Verification involves confirming or
denying a person's claimed identity. (Are
you who you claim to be?).
Using Biometrics for identification, for
example, DNA evidence or fingerprints,
has been used for many years. It is widely
accepted as being credible and accurate.
Biometrics is now being used for
verification as part of security systems.
Verification has been part of security
systems for many years. We have been
asked for physical items of identification
like a license or passport. We have also
been asked for things we know to verify
our identity, like a pin number or a maiden
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 3/17
name. Unfortunately, security systems that
use physical forms of identification and
personal information can be easily
"fooled". It is the increasing need for more
security that has enabled biometrics to
evolve to its current state of development.
Biometrics determines who we are using
our distinct physical features.
PARTS OF BIOMETRICS
Biometric systems consist of three parts:
1.Scanner:
Scans the anatomy being used.
2.Software:
The software which gathers the
information and converts it to digital form
3.Database:
The database which compares the current
data to a stored database and determines
authenticity and identification.
CLASSIFICATION
Biometrics is classified into two
categories:
1.Physiological
Physiological biometrics measures the
distinct traits that people have, usually (but
not always or entirely) dictated by their
genetics. They are based on measurements
and data derived from direct measurement
of a part of the human body.
Physiological Biometric Applications:
• Fingerprints
• Retina Scans
• Hand Geometry
• Facial Patterns
• DNA
2.Behavioral
Behavioral biometrics measure the distinctactions that humans take, which are
generally very hard to copy from one
person to another. They measure
characteristics of the human body
indirectly.
Behavioral Biometric Applications:
• Speaker Recognition
• Gait
• Signature
• Keystroke
Physiological
1. Fingerprints
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 4/17
Fingerprinting takes an image
(either using ink or a digital scan) of a
person's fingertips and records its
characteristics. Most developed and widely
accepted type of biometric. It uses the
patterns of whorls, arches, and loops along
with patterns of ridges, furrows and
minutise. Fingerprints can be compared to
a database to determine identification.
Fingerprinting was first used in this
country by the New York civil service
since 1902 and was first discussed in the
scientific community and proposed as a
means of identification in 1880. No twohumans have ever been found to have the
same fingerprint. Even identical twins, that
are twins that resulted from one embryo
that divided to make two offspring, have
unique fingerprints. In 2000, electronic
fingerprints were introduced for
verification of the user during computer
login procedures.
Pros:
• No image of the fingerprint is
actually created, therefore no
means to "steal fingerprints".
• Takes less than 5 seconds.
• To prevent fake fingers, many
systems also measure blood flow.
• 1:5000 accuracy for a single finger
•Using multiple fingers increasesaccuracy exponentially.
• Small storage space required for
biometric template.
• Fingerprinting would be a very
good form of biometrics to use in
schools as fingerprint patterns are
set at birth and do not change asthe individual grows.
Cons:
• Fingerprints can be "lifted" very
easily by using a dusting powder
and tape as is done routinely in
crime scene investigations.
Scanners can easily be fooled by
using a similar technique.
2. Retina Scan:
The retina is the thin layer of cells at the
back of the eyeball that transmits images
into neural signals. The retina is very rich
in blood vessels and each individual has a
distinct pattern of blood vessels on their
retina. Retina scans can compare the
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 5/17
patterns to a known data base to confirm
identification of an individual.
Retinal scans take 10 to 15
seconds. Development on retinal scans
began in the 1930's. The first device
available for commercial use was in 1984
and was called the Eyedentification 7.5
personal identification unit, developed by
EyeDentify founded in 1976.
Pros:
• Retina scan devices are one of the
most accurate biometric
applications available today as
retinal patterns remain constant
throughout an individual's life.
Cons:
• Requires user to remove glasses to
ensure that the user can place their
eye close to the device.
• General public perceives the laser
as being potentially harmful to the
eye.
• Hardware is expensive.
• System not easy to use.
3.Voice:
With voice recognition, the user speaks
into a microphone his password or access
phrase. Verification time is approximately
5 seconds.
Based on physiological and
behavioral differences in speech as the
sound waves are produced. Physiological
differences arise from differences in
human's vocal tract. These includestructural differences in the pharynx, oral
and nasal cavities as well as the shape of
the vocal cords.
Pros:
• Easy to use and takes
approximately 5 seconds for
verification.
Cons:
• To prevent compromise of the
system by use of a recorded voice,
the majority of voice devices
require the high and low
frequencies of the sound to match.
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 6/17
This is difficult to recreate well for
many recording instruments.
4.Facial:
Facial recognition began development in
the late 1980's. Facial recognition analyzes
facial images using a video camera. It
measures structure like distance between
the eyes, between the eyes and nose, eyes
and mouth, nose and mouth, etc. Users
stand approximately 3 feet from the
camera and are often required to blink or
smile to prevent successful use of fake
faces.
Pros:
• Non-intrusive, users stand several
feet away. Not required to wait for
long periods of time.
Cons:
• Non-intrusive nature gives a sense
of "Big Brother".
5. Hand Geometry:
Involves the measurement and
analysis of the shape of an individual's
hand. User places hand on a metal
template. Process takes approximately 5
seconds. Has been used for 20 years.
Pros:
• Special hardware required can be
easily integrated with other
biometric systems like
fingerprinting.
• Easy for users to use the system.
• Resistant to tampering since the
effort required to make a fake hand
is not worth the effort.
Cons:
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 7/17
• Hand geometry is not unique to an
individual, therefore possible to
have duplicate readings.
•
Hardware is costly and takes upmuch space.
• Injuries to hands cause problems
with the system.
6. Gait:
Gait is the peculiar way one walks
and is a complex spatio-temporal
behavioral biometrics. Gait is not
supposed to be unique to each individual,
but is sufficiently characteristic to allow
identity authentication. Gait is a behavioral biometric and may not stay invariant
especially over a large period of time, due
to large fluctuations of body weight, major
shift in the body weight (e.g., waddling
gait during pregnancy, major injuries
involving joints or brain (e.g., cerebellar
lesions in Parkinson disease), or due to
inebriety (e.g., drunken gait).
Humans are quite adept at
recognizing a person at a distance from his
gait. Although, the characteristic gait of a
human walk has been well researched in
the Department of Immigration and
Naturalization in the United States
specifically requests photographs of
individuals with clearly visible right ear.
Biomechanics community to detect
abnormalities in lower extremity joints, the
use of gait for identification purposes isvery recent. Typically, gait features are
derived from an analysis of video-
sequence footage of a walking person and
consist of characterization of several
different movements of each articulate
joint. Currently, there do not exist any
commercial systems for performing gait-
based authentication. The method of input
acquisition for gait is not different from
that of acquiring facial pictures, and hence
gait may be an acceptable biometric. Since
gait determination involves processing of
video, it is compute and input intensive.
7. DNA:
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 8/17
DNA (DeoxyriboNucleic Acid) is
the one-dimensional ultimate unique code
for one's individuality - except for the fact
that identical twins have the identical
DNA pattern. It is, however, currently
used mostly in the context of forensic
applications for identification.
Three issues limit the utility of this
biometrics for other applications:
(i) Contamination and sensitivity: It is
easy to steal a piece of DNA from an
unsuspecting subject to be subsequently
abused for an ulterior purpose;
(ii ) Automatic real-time identification
issues: The present technology for genetic
matching is not geared for online
unobtrusive identifications. Most of the
human DNA is identical for the entire
human species and only some relatively
small number of specific locations
(polymorphic loci) on DNA exhibit
individual variation. These variations are
manifested either in the number of
repetitions of a block of base sequence
(length polymorphism) or in the minor non-functional perturbations of the base
sequence (sequence polymorphism). The
processes involved in DNA based personal
identification determine whether two DNA
samples originate from the same/different
individual(s) based on the distinctive
signature at one or more polymorphic loci.
A major component of these processes
now exist in the form of cumbersome
chemical methods (wet processes)
requiring an expert's skills. There does not
seem to be any effort directed at a
complete automation of all the processes.
iii) Privacy issues: Information about
susceptibilities of a person to certain
diseases could be gained from the DNA
pattern and there is a concern that the
unintended abuse of genetic code
information may result in discrimination ine.g., hiring practices.
Important points to observe in DNA:
• Only 2-3% of the DNA sequence
represents the known genetic
material.
• Almost 70% of the sequence is
composed of non-coding regions,
i.e. we do not know the function of
these regions.
• Almost 30% of the sequence is
composed of non-coding repetitive
DNA, and only 1/3 is tandemly
repetitive, the rest (2/3) is
randomly repetitive.
8. Signature and Acoustic Emissions:
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 9/17
The way a person signs her name is known
to be a characteristic of that individual.
Although signatures require contact and
effort with the writing instrument, they
seem to be acceptable in many
government, legal, and commercial
transactions as a method of personal
authentication.
Signatures are a behavioral biometric, evolve over a period of time
and are influenced by physical and
emotional conditions of the signatories.
Signatures of some people vary a lot: even
the successive impressions of their
signature are significantly different.
Further, the professional forgers can
reproduce signatures to fool the unskilled
eye. Although, the human experts can
discriminate genuine signatures from the
forged ones, modeling the invariance in
the signatures and automating signature
recognition process are challenging. There
are two approaches to signature
verification: static and dynamic. In static
signature verification, only geometric
(shape) features of the signature are used
for authenticating an identity.
Typically, the signature
impressions are normalized to a known
size and decomposed into simple
components (strokes). The shapes and
relationships of strokes are used as
features. In dynamic signature verification,
not only the shape features are used for
authenticating the signature but thedynamic features like acceleration,
velocity, and trajectory profiles of the
signature are also employed. The signature
impressions are processed as in a static
signature verification system. Invariants of
the dynamic features augment the static
features, making forgery difficult since the
forger has to not only know the impression
of the signature but also the way the
impression was made. A related
technology is authentication of an identity
based on the characteristics of the acoustic
emissions emitted during a signature
scribble.
9. Keystroke Dynamics:
Keystroke dynamics is an automated
method of examining an individual’s
keystrokes on a keyboard. This technology
examines such dynamics as speed and
pressure, the total time taken to type
particular words, and the time elapsed
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 10/17
between hitting certain keys. This
technology’s algorithms are still being
developed to improve robustness and
distinctiveness. One potentially useful
application that may emerge is computer
access, where this biometric could be used
to verify the computer user’s identity
continuously.
DETAILED DISCRIPITION ON FACE
RECOGNITION:
Face recognition:
As a biometric, facial recognition is a form
of computer vision that uses faces to
attempt to identify a person or verify a
person’s claimed identity. Regardless of
specific method used, facial recognition is
accomplished in a five step process.
1. First, an image of the face is acquired.
This acquisition can be accomplished by
digitally scanning an existing photograph
or by using an electro-optical camera to
acquire a live picture of a subject. As
video is a rapid sequence of individual still
images, it can also be used as a source of
facial images.
2. Second, software is employed to detect
the location of any faces in the acquired
image. This task is difficult, and often
generalized patterns of what a face
“looks like” (two eyes and a
mouth set in an oval shape) are
employed to pick out the faces.
3. Once the facial detection
software has targeted a face, it
can be analyzed. As noted in slide
three, facial recognition analyzes
the spatial geometry of
distinguishing features of the
face. Different vendors use
different methods to extract the
identifying features of a face.
Thus, specific details on the
methods are proprietary. The
most popular method is called
Principle Components Analysis
(PCA), which is commonly
referred to as the eigenface
method. PCA has also been
combined with neural networks
and local feature analysis in
efforts to enhance its
performance. Template
generation is the result of the
feature extraction process. A
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 11/17
template is a reduced set of data
that represents the unique
features of an enrollee’s face. It is
important to note that becausethe systems use spatial geometry
of distinguishing facial features,
they do not use hairstyle, facial
hair, or other similar factors.
4. The fourth step is to compare
the template generated in step
three with those in a database of
known faces. In an identification
application, this process yields
scores that indicate how closely
the generated template matches
each of those in the database. In
a verification application, the
generated template is only
compared with one template in
the database – that of the claimed
identity.
5. The final step is determining
whether any scores produced in
step four are high enough to
declare a match. The rules
governing the declaration of a
match are often configurable by
the end user, so that he or she
can determine how the facial
recognition system should behave
based on security and operational
considerations.
This graphic depicts a notional
facial recognition surveillance
system. Read clockwise from thelower left-hand corner, this
system identifies and locates
“targets” (e.g., criminals, suspect
terrorists, missing children)
through a networked system of
surveillance cameras (or CCTV).
Video streams are sent over anetwork to a central control
facility (e.g., “Control Room”). At
that central facility, computers
find faces in the video, and then
attempt to find a match in a
database of target individuals. If a
probable match is found, thesystem alerts an officer; it
presents him with the image of
the suspected match, as well as
the image of the individual in the
database. This verification step
uses trained officers to ensure
that false alarms generated by
the system are caught and
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 12/17
recorded. If the officer decides
that the match is not a false
alarm, he forwards the alert to
officers on patrol, who are in thevicinity of where the original
camera filmed the suspect.
People are generally very good at
recognizing faces that they know.
However, people experience
difficulties when they perform
facial recognition in surveillance
or watch post scenario. Several
factors account for these
difficulties: most notably, humans
have a hard time recognizing
unfamiliar faces. Combined with
relatively short attention spans, it
is difficult for humans to pick out
unfamiliar faces. Considerable
evidence supports this claim. For
example, in a British study,
trained supermarket cashiers
were tested on their ability to
screen shoppers using credit
cards that included a photograph
of the card owner. Each shopper
was issued four cards: one with arecent picture of the shopper, one
that included minor modifications
to the shopper’s hairstyle, facial
hair or accessories (e.g., glasses,
hat), another card with a
photograph of a person similar in
appearance to the shopper, and
the last card with a photograph of
a person who was only of the
same sex and race as the
shopper. When the various cards
were presented to the checkout
clerks, more than half of the
fraudulent cards were accepted.
The breakdown was as follows: 34
percent of the cards that did not
look like the shopper were
accepted, 14 percent of the cards
where the appearance had been
altered were accepted, and 7
percent of the unchanged cards
were rejected by the clerks. In
addition to unfamiliar face
recognition problems, the ability
of human beings to detect critical
signals drops rapidly from the
start of a task, and stabilizes at a
significantly lower level within 25
to 35 minutes. Thus the ability of
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 13/17
people to focus their attention
drops significantly after only half
an hour.
People are generally very good atrecognizing faces that they know.
However, people experience difficulties
when they perform facial recognition in
surveillance or watch post scenario.
Several factors account for these
difficulties: most notably, humans have a
hard time recognizing unfamiliar faces.
Combined with relatively short attention
spans, it is difficult for humans to pick out
unfamiliar faces. Considerable evidence
supports this claim. For example, in a
British study, trained supermarket cashiers
were tested on their ability to screen
shoppers using credit cards that included a
photograph of the card owner. Each
shopper was issued four cards: one with a
recent picture of the shopper, one that
included minor modifications to the
shopper’s hairstyle, facial hair or
accessories (e.g ., glasses, hat), another
card with a photograph of a person similar
in appearance to the shopper, and the last
card with a photograph of a person who
was only of the same sex and race as the
shopper. When the various cards were
presented to the checkout clerks, more
than half of the fraudulent cards were
accepted. The breakdown was as follows:
34 percent of the cards that did not look
like the shopper were accepted, 14 percent
of the cards where the appearance had
been altered were accepted, and 7 percent
of the unchanged cards were rejected by
the clerks. In addition to unfamiliar face
recognition problems, the ability of human
beings to detect critical signals drops
rapidly from the start of a task, and
stabilizes at a significantly lower level
within 25 to 35 minutes. Thus the ability
of people to focus their attention drops
significantly after only half an hour.
Machines also experience difficulties
when they perform facial recognition in
surveillance or watch post scenario. Dr.
James L. Wayman, a leading biometrics
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 14/17
expert, has explained that performing
facial recognition processes with relatively
high fidelity and at long distances remains
technically challenging for automated
systems. At the most basic level, detecting
whether a face is present in a given
electronic photograph is a difficult
technical problem. Dr. Wayman has noted
that subjects should ideally be
photographed under tightly controlled
conditions. For example, each subject
should look directly into the camera and
fill the area of the photo for an automated
system to reliably identify the individual
or even detect his face in the photograph.
Thus, while the technology for facial
recognition systems shows promise, it is
not yet considered fully mature. The
“Facial Recognition Vendor Test 2000”
study makes clear that the technology is
not yet perfected. This comprehensive
study of current facial recognition
technologies, sponsored by the
Department of Defense (DoD)
Counterdrug Technology Development
Program Office, the Defense Advanced
Research Projects Agency (DARPA), and
the National Institute of Justice, showed
that environmental factors such as
differences in camera angle, direction of
lighting, facial expression, and other
parameters can have significant effects on
the ability of the systems to recognize
individuals.
By controlling a person’s facial
expression, as well as his distance from the
camera, the camera angle, and the scene’s
lighting, a posed image minimizes the
number of variables in a photograph. This
control allows the facial recognition
software to operate under near ideal
conditions – greatly enhancing its
accuracy. Similarly, using a human
operator to verify the system’s results
enhances performance because the
operator can detect machine-generated
false alarms.
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 15/17
An “obvious” point that needs stating: The
better the quality of the captured image
and the database images, the better the
facial recognition system will perform.
The “facetrap” triangle above
demonstrates this point, with respect to
acquiring high-quality images of the
target’s face. It is difficult to acquire an
image if the authorities only know that a
suspect might be at an airport west of the
Mississippi River. It is easier to capture
the image at a facetrap. For example, a
surveillance camera can more easily
capture images of people at the check-in
counter. Sometimes facetraps can be
designed to take advantage of people’s
inclinations. For example, a person going
up an escalator will naturally look at a red
flashing light above a clock at the top of
the escalator. A surveillance camera
located there can easily capture an image;
the face has been trapped. A camera
located at a metal detector also takes
advantage of a facetrap. The best facetrap
is the one shown at the apex of the triangle
—an image captured under tightly
controlled conditions.
The following factors need to be
considered with respect to testing and
evaluation of facial recognition systems:
1. Testing should be conducted by
independent organizations that will not
reap any benefits should one system
outperform another (i.e. no conflicts of
interest involved). The Facial Recognition
Vendor Test (FVRT) testing which
government agencies sponsor is likely to be very objective.
2. The test philosophy must be considered.
For example, the FVRT tries to make the
test neither too difficult nor too easy, as it
does not want all the systems’
performance to cluster at one end of the
spectrum. The FVRT also wants to
distinguish performance of systems and
give feedback to designers for
improvement. But a drawback here is that
real life data does not present itself this
way. Performance in real life may very
well prove that none of the systems are
useful.
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 16/17
3. Vendors and developers should not
know test data beforehand; otherwise, they
may be tempted to fine-tune their
technology’s performance to the specific
test data. Performance data that has been
fine-tuned to specific test data is not
representative of the general performance
of the technology being tested.
4. Testing and evaluation should be
repeatable. That is, statistically similar
results should be able to be reproduced.
In the final analysis, real life
deployments will be the ultimate tests of
FR systems. For now the jury is still out on
the effectiveness of facial recognition
systems, however, the technology is
improving. Facial recognition systems
may yet become a part of our daily lives as
they improve and if they become more
acceptable, much as CCTV or surveillance
camera systems have become.
Leading products in biometrics:
Biometric is a new but promising
technology and therefore a number of
companies have appeared in the market in
a very short period of time. Some of those
products are:
Conclusion:
Biometrics is the most sought after branch today and is widely renowned for
its various applications in different fields
especially for the sake of security. It is
predominantly used to identify the
individuals in order to know their identity
by using the various techniques that were
just explained above in order to create
better security. The advances in accuracy
and usability and decreasing cost have
made the biometric technology a secure,
affordable and cost effective way of
identifying individuals. Thus helping the
people to create more safe and secure
environment especially in today’s
hazardous situations.
8/8/2019 Biometrics ou
http://slidepdf.com/reader/full/biometrics-ou 17/17