BIOMETRICS

KAMALA INSTITUTE OF TECHNOLOGY & SCIENCE

SINGAPUR, HUZURABAD
2008-2009
ABSTRACT INTRODUCTION

Biometrics refers to the automatic Definition: Biometrics is the process by

identification of a person based on his/her which distinguishing human anatomy is
physiological or behavioral characteristics. used for identification and verification.
This method of identification is preferred Biometric systems are pattern recognition
over traditional methods involving systems which are automated methods of
password’s and PIN numbers. determining the authenticity of a specific
physiological or behavioral characteristic
Using Biometrics you can unlock
possessed by the user to determine or
your houses, withdrawing money from a
verify identify. The word biometrics
bank with just a blink of an eye, a tap of
comes from the Greek words bio -
your finger or by just showing your face.
meaning life, and metric - meaning to
Fingerprint recognition is most widely
measure.
accepted biometric among the technology
being used today. Biometrics in Face Identification is determining who a
recognition has received a surge of person is. One has to establish a person's
attention since of disaster of 11/9 for its identity (Who am I?).
ability to identify known terrorists and
criminals. Verification involves confirming or
denying a person's claimed identity. (Are
First usage:- you who you claim to be?).

First used in China in the 14th Using Biometrics for identification, for
century -- merchants stamped hand and example, DNA evidence or fingerprints,
foot prints of children on paper with ink to has been used for many years. It is widely
distinguish one child from another. In accepted as being credible and accurate.
Europe, Richard Edward Henry developed Biometrics is now being used for
fingerprinting for Scotland Yard in the verification as part of security systems.
later half of the 19th century. 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
name. Unfortunately, security systems that of a part of the human body.
use physical forms of identification and
personal information can be easily Physiological Biometric Applications:
"fooled". It is the increasing need for more
• Fingerprints
security that has enabled biometrics to
• Retina Scans
evolve to its current state of development.
• Hand Geometry
Biometrics determines who we are using
• Facial Patterns
our distinct physical features.
• DNA
PARTS OF BIOMETRICS
2.Behavioral
Biometric systems consist of three parts: Behavioral biometrics measure the distinct
actions that humans take, which are
1.Scanner:
generally very hard to copy from one
Scans the anatomy being used.
person to another. They measure
2.Software:
characteristics of the human body
The software which gathers the
indirectly.
information and converts it to digital form
3.Database: Behavioral Biometric Applications:
The database which compares the current
data to a stored database and determines • Speaker Recognition

authenticity and identification. • Gait

• Signature
• Keystroke
CLASSIFICATION

Biometrics is classified into two

categories:
Physiological
1.Physiological
Physiological biometrics measures the
distinct traits that people have, usually (but
1. Fingerprints
not always or entirely) dictated by their
genetics. They are based on measurements
and data derived from direct measurement
 • To prevent fake fingers, many
systems also measure blood flow.
• 1:5000 accuracy for a single finger
• Using multiple fingers increases
Fingerprinting takes an image accuracy exponentially.
(either using ink or a digital scan) of a • Small storage space required for
person's fingertips and records its biometric template.
characteristics. Most developed and widely
• Fingerprinting would be a very
accepted type of biometric. It uses the
good form of biometrics to use in
patterns of whorls, arches, and loops along
schools as fingerprint patterns are
with patterns of ridges, furrows and
set at birth and do not change as
minutise. Fingerprints can be compared to
the individual grows.
a database to determine identification.
Cons:

Fingerprinting was first used in this

• Fingerprints can be "lifted" very
country by the New York civil service
easily by using a dusting powder
since 1902 and was first discussed in the
and tape as is done routinely in
scientific community and proposed as a
crime scene investigations.
means of identification in 1880. No two
Scanners can easily be fooled by
humans have ever been found to have the
using a similar technique.
same fingerprint. Even identical twins, that
are twins that resulted from one embryo 2. Retina Scan:
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: The retina is the thin layer of cells at the

back of the eyeball that transmits images
• No image of the fingerprint is into neural signals. The retina is very rich
actually created, therefore no in blood vessels and each individual has a
means to "steal fingerprints". distinct pattern of blood vessels on their
• Takes less than 5 seconds. retina. Retina scans can compare the
patterns to a known data base to confirm 3.Voice:
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.
With voice recognition, the user speaks
Pros:
into a microphone his password or access

• Retina scan devices are one of the phrase. Verification time is approximately

most accurate biometric 5 seconds.

applications available today as Based on physiological and

retinal patterns remain constant behavioral differences in speech as the
throughout an individual's life. sound waves are produced. Physiological

Cons: differences arise from differences in

human's vocal tract. These include
• Requires user to remove glasses to structural differences in the pharynx, oral
ensure that the user can place their and nasal cavities as well as the shape of
eye close to the device. the vocal cords.

• General public perceives the laser Pros:

as being potentially harmful to the
• Easy to use and takes
eye.
approximately 5 seconds for
• Hardware is expensive. verification.

• System not easy to use. 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.
 This is difficult to recreate well for • Non-intrusive nature gives a sense
many recording instruments. of "Big Brother".

4.Facial: 5. Hand Geometry:

Facial recognition began development in Involves the measurement and

the late 1980's. Facial recognition analyzes analysis of the shape of an individual's
facial images using a video camera. It hand. User places hand on a metal
measures structure like distance between template. Process takes approximately 5
the eyes, between the eyes and nose, eyes seconds. Has been used for 20 years.
and mouth, nose and mouth, etc. Users
Pros:
stand approximately 3 feet from the
camera and are often required to blink or • Special hardware required can be
smile to prevent successful use of fake easily integrated with other
faces. biometric systems like
fingerprinting.
Pros:
• Easy for users to use the system.
• Non-intrusive, users stand several • Resistant to tampering since the
feet away. Not required to wait for effort required to make a fake hand
long periods of time. is not worth the effort.

Cons:
Cons:
 • Hand geometry is not unique to an Humans are quite adept at
individual, therefore possible to recognizing a person at a distance from his
have duplicate readings. gait. Although, the characteristic gait of a
• Hardware is costly and takes up human walk has been well researched in
much space. the Department of Immigration and

• Injuries to hands cause problems Naturalization in the United States

with the system. specifically requests photographs of

individuals with clearly visible right ear.

6. Gait: Biomechanics community to detect

abnormalities in lower extremity joints, the
use of gait for identification purposes is
very 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
Gait is the peculiar way one walks
that of acquiring facial pictures, and hence
and is a complex spatio-temporal
gait may be an acceptable biometric. Since
behavioral biometrics. Gait is not
gait determination involves processing of
supposed to be unique to each individual,
video, it is compute and input intensive.
but is sufficiently characteristic to allow
identity authentication. Gait is a behavioral
biometric and may not stay invariant 7. DNA:
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).
 DNA (DeoxyriboNucleic Acid) is now exist in the form of cumbersome
the one-dimensional ultimate unique code chemical methods (wet processes)
for one's individuality - except for the fact requiring an expert's skills. There does not
that identical twins have the identical seem to be any effort directed at a
DNA pattern. It is, however, currently complete automation of all the processes.
used mostly in the context of forensic iii) Privacy issues: Information about
applications for identification. susceptibilities of a person to certain
Three issues limit the utility of this diseases could be gained from the DNA
biometrics for other applications: pattern and there is a concern that the

(i) Contamination and sensitivity: It is unintended abuse of genetic code

easy to steal a piece of DNA from an information may result in discrimination in

unsuspecting subject to be subsequently e.g., hiring practices.

abused for an ulterior purpose; Important points to observe in DNA:

(ii) Automatic real-time identification • Only 2-3% of the DNA sequence

issues: The present technology for genetic represents the known genetic
matching is not geared for online material.
unobtrusive identifications. Most of the
• Almost 70% of the sequence is
human DNA is identical for the entire
composed of non-coding regions,
human species and only some relatively
i.e. we do not know the function of
small number of specific locations
these regions.
(polymorphic loci) on DNA exhibit
• Almost 30% of the sequence is
individual variation. These variations are
composed of non-coding repetitive
manifested either in the number of
DNA, and only 1/3 is tandemly
repetitions of a block of base sequence
repetitive, the rest (2/3) is
(length polymorphism) or in the minor
randomly repetitive.
non-functional perturbations of the base
sequence (sequence polymorphism). The 8. Signature and Acoustic Emissions:
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
 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,
The way a person signs her name is known not only the shape features are used for
to be a characteristic of that individual. authenticating the signature but the
Although signatures require contact and dynamic features like acceleration,
effort with the writing instrument, they velocity, and trajectory profiles of the
seem to be acceptable in many signature are also employed. The signature
government, legal, and commercial impressions are processed as in a static
transactions as a method of personal signature verification system. Invariants of
authentication. the dynamic features augment the static

Signatures are a behavioral features, making forgery difficult since the

biometric, evolve over a period of time forger has to not only know the impression

and are influenced by physical and of the signature but also the way the

emotional conditions of the signatories. impression was made. A related

Signatures of some people vary a lot: even technology is authentication of an identity

the successive impressions of their based on the characteristics of the acoustic

signature are significantly different. emissions emitted during a signature

Further, the professional forgers can scribble.

reproduce signatures to fool the unskilled

eye. Although, the human experts can
9. Keystroke Dynamics:
discriminate genuine signatures from the
Keystroke dynamics is an automated
forged ones, modeling the invariance in
method of examining an individual’s
the signatures and automating signature
keystrokes on a keyboard. This technology
recognition process are challenging. There
examines such dynamics as speed and
are two approaches to signature
pressure, the total time taken to type
verification: static and dynamic. In static
particular words, and the time elapsed
between hitting certain keys. This images, it can also be used as a source of
technology’s algorithms are still being facial images.
developed to improve robustness and
distinctiveness. One potentially useful 2. Second, software is employed to detect
application that may emerge is computer the location of any faces in the acquired
access, where this biometric could be used image. This task is difficult, and often
to verify the computer user’s identity generalized patterns of what a face
continuously. “looks like” (two eyes and a

DETAILED DISCRIPITION ON FACE mouth set in an oval shape) are

RECOGNITION: employed to pick out the faces.

Face recognition:
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

As a biometric, facial recognition is a form identifying features of a face.

of computer vision that uses faces to Thus, specific details on the

attempt to identify a person or verify a methods are proprietary. The

person’s claimed identity. Regardless of most popular method is called

specific method used, facial recognition is Principle Components Analysis

accomplished in a five step process. (PCA), which is commonly

referred to as the eigenface

1. First, an image of the face is acquired. method. PCA has also been

This acquisition can be accomplished by combined with neural networks

digitally scanning an existing photograph and local feature analysis in
or by using an electro-optical camera to efforts to enhance its
acquire a live picture of a subject. As performance. Template
video is a rapid sequence of individual still generation is the result of the
feature extraction process. A
template is a reduced set of data
that represents the unique
features of an enrollee’s face. It is
important to note that because
the systems use spatial geometry
of distinguishing facial features,
they do not use hairstyle, facial
hair, or other similar factors.

This graphic depicts a notional

4. The fourth step is to compare facial recognition surveillance
the template generated in step system. Read clockwise from the
three with those in a database of lower left-hand corner, this
known faces. In an identification system identifies and locates
application, this process yields “targets” (e.g., criminals, suspect
scores that indicate how closely terrorists, missing children)
the generated template matches through a networked system of
each of those in the database. In surveillance cameras (or CCTV).
a verification application, the Video streams are sent over a
generated template is only network to a central control
compared with one template in facility (e.g., “Control Room”). At
the database – that of the claimed that central facility, computers
identity. find faces in the video, and then
5. The final step is determining attempt to find a match in a
whether any scores produced in database of target individuals. If a
step four are high enough to probable match is found, the
declare a match. The rules system alerts an officer; it
governing the declaration of a presents him with the image of
match are often configurable by the suspected match, as well as
the end user, so that he or she the image of the individual in the
can determine how the facial database. This verification step
recognition system should behave uses trained officers to ensure
based on security and operational that false alarms generated by
considerations. the system are caught and
recorded. If the officer decides screen shoppers using credit
that the match is not a false cards that included a photograph
alarm, he forwards the alert to of the card owner. Each shopper
officers on patrol, who are in the was issued four cards: one with a
vicinity of where the original recent picture of the shopper, one
camera filmed the suspect. 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
People are generally very good at clerks, more than half of the
recognizing faces that they know. fraudulent cards were accepted.
However, people experience The breakdown was as follows: 34
difficulties when they perform percent of the cards that did not
facial recognition in surveillance look like the shopper were
or watch post scenario. Several accepted, 14 percent of the cards
factors account for these where the appearance had been
difficulties: most notably, humans altered were accepted, and 7
have a hard time recognizing percent of the unchanged cards
unfamiliar faces. Combined with were rejected by the clerks. In
relatively short attention spans, it addition to unfamiliar face
is difficult for humans to pick out recognition problems, the ability
unfamiliar faces. Considerable of human beings to detect critical
evidence supports this claim. For signals drops rapidly from the
example, in a British study, start of a task, and stabilizes at a
trained supermarket cashiers significantly lower level within 25
were tested on their ability to to 35 minutes. Thus the ability of
people to focus their attention card with a photograph of a person who
drops significantly after only half was only of the same sex and race as the
an hour. 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
People are generally very good at
the clerks. In addition to unfamiliar face
recognizing faces that they know.
recognition problems, the ability of human
However, people experience difficulties
beings to detect critical signals drops
when they perform facial recognition in
rapidly from the start of a task, and
surveillance or watch post scenario.
stabilizes at a significantly lower level
Several factors account for these
within 25 to 35 minutes. Thus the ability
difficulties: most notably, humans have a
of people to focus their attention drops
hard time recognizing unfamiliar faces.
significantly after only half an hour.
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 Machines also experience difficulties

accessories (e.g., glasses, hat), another when they perform facial recognition in

card with a photograph of a person similar surveillance or watch post scenario. Dr.

in appearance to the shopper, and the last James L. Wayman, a leading biometrics
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 By controlling a person’s facial
should look directly into the camera and expression, as well as his distance from the
fill the area of the photo for an automated camera, the camera angle, and the scene’s
system to reliably identify the individual lighting, a posed image minimizes the
or even detect his face in the photograph. number of variables in a photograph. This
Thus, while the technology for facial control allows the facial recognition
recognition systems shows promise, it is software to operate under near ideal
not yet considered fully mature. The conditions – greatly enhancing its
“Facial Recognition Vendor Test 2000” accuracy. Similarly, using a human
study makes clear that the technology is operator to verify the system’s results
not yet perfected. This comprehensive enhances performance because the
study of current facial recognition operator can detect machine-generated
technologies, sponsored by the false alarms.
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.
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
The following factors need to be
suspect might be at an airport west of the
considered with respect to testing and
Mississippi River. It is easier to capture
evaluation of facial recognition systems:
the image at a facetrap. For example, a
1. Testing should be conducted by
surveillance camera can more easily
independent organizations that will not
capture images of people at the check-in
reap any benefits should one system
counter. Sometimes facetraps can be
outperform another (i.e. no conflicts of
designed to take advantage of people’s
interest involved). The Facial Recognition
inclinations. For example, a person going
Vendor Test (FVRT) testing which
up an escalator will naturally look at a red
government agencies sponsor is likely to
flashing light above a clock at the top of
be very objective.
the escalator. A surveillance camera
2. The test philosophy must be considered.
located there can easily capture an image;
For example, the FVRT tries to make the
the face has been trapped. A camera
test neither too difficult nor too easy, as it
located at a metal detector also takes
does not want all the systems’
advantage of a facetrap. The best facetrap
performance to cluster at one end of the
is the one shown at the apex of the triangle
spectrum. The FVRT also wants to
—an image captured under tightly
distinguish performance of systems and
controlled conditions.
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.
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 Conclusion:

results should be able to be reproduced. Biometrics is the most sought after

In the final analysis, real life branch today and is widely renowned for
deployments will be the ultimate tests of its various applications in different fields
FR systems. For now the jury is still out on especially for the sake of security. It is
the effectiveness of facial recognition predominantly used to identify the
systems, however, the technology is individuals in order to know their identity
improving. Facial recognition systems by using the various techniques that were
may yet become a part of our daily lives as just explained above in order to create
they improve and if they become more better security. The advances in accuracy
acceptable, much as CCTV or surveillance and usability and decreasing cost have
camera systems have become. made the biometric technology a secure,
affordable and cost effective way of
Leading products in biometrics: identifying individuals. Thus helping the

Biometric is a new but promising people to create more safe and secure

technology and therefore a number of environment especially in today’s

companies have appeared in the market in hazardous situations.

a very short period of time. Some of those

products are: