Biometric Authentication System: Tools and ... - rs publication

nauseatingcynicalSécurité

22 févr. 2014 (il y a 3 années et 10 mois)

103 vue(s)



Page
150


I
NTERNATIONAL
J
OURNAL OF
C
OMPUTER
A
PPLICATION

I
SSUE
2,

V
OLUME
2

(APRIL

2012)





ISSN:

2250
-
1797

Biometric

Authentication

System
: Tools and Techniques


1
Ishpreet Singh Virk,
2
Raman Maini

1
Research Scholar, U
niversity College of Engineering, Punjabi University, Patiala

ishvirk55@gmail.com

2
Associate Professor, University College of Engineering, Punjabi

University, Patiala

research_raman@yahoo.com


_________________________________________________________________________
_______________________________________


Abstract

In today world of technology, machines are replacing human being in every aspect of life, so it is incr
easing need of the security concern to
automate the

different surveillance techniques and

authentication
checking
of the users i.e.

to reject unauthorized persons.
This paper discuss

the new technolo
gy

in the market: the biometric technology that is an ide
ntification and verification system based on measurements of
biological traits.

The paper also provides an overview of the different biometric available with one advantages and disadvantages and their
difference by the implementation of the biometric techn
iques.

Biometric techniques will affect privacy for all people, taking both information
privacy and personal integrity into consideration.

The increase in

the flow of people, goods, services and capital and it increase the spreading
of information across I
nternet.
New

dangers are created and an increasing demand for safety and security in the public and private sectors is
driving research in the field of automated recognition of individuals by using behavioral and biological characteristics
.


Although a
lot of work has been done in the field of Biometric, particularly in the field of fingerprint recognition but highly secured
and fast
systems are still to be achieved. Further we can use multimodal biometric system, which uses more than one biometric trait

such as
fingerprint with voice and
etc. Main

concern in any biometric system is FRR (False reject rat
e) and FAR (False accept rate).


_________________________________________________________________________
____________________________________

C
orrponding
author :
Ishpreet Singh Virk

1
: Introduction

The word biometrics derived from “bio” that means life and “metric” that means measurement, in other word
s is the study of methods to uniquely
recognize human traits of each person. The study of automated identification, by use of physical or behavio
ral traits is called biometrics [1
].

The application of biometric is security. Biometric technologies are posit
ioning themselves as the foundation for many highly secure identification
and personal verification solutions. Areas that will benefit from biometric technologies include network security infrastruct
ures, government IDs,
secure electronic banking, investin
g and financial transactions, wireless communications, retail, and health and social services. Highly secure and
trustworthy electronic commerce, for example, will be essential to the healthy growth of the global economy. Many biometric t
echnology provider
s
are already delivering biometric authentication for a variety of web
-
based and client/server based applications to meet these and other needs.

Biometric system can be either an 'identification' system or a 'verification'

(authentication) sy
stem, which are defined below

[2
].



Figure 1: Biometric Solutions


Identification
-

One to Many: Biometrics can be used to determine a person's identity even without his knowledge or consent. For example, scan
ning
a crowd with a camera and using face recognition tech
nology, one can determine matches against a known database.


Verification
-

One to One: Biometrics can also be used to verify a person's identity. For example, one can grant physical access to a secure

area in a
building by using finger scans or can grant
access to a bank account at

an ATM by using retinal scan [1
].


The word biometric can also be associated to a larger set of information that a person own, the three categories are: Who you

are, what you have and
what

you know.

Among the various biometric technologies being considered, the attributes which satisfy the above requirements are fingerprin
t,
hand geometry, palm print, facial features, iris, voice, vein, retina, patterns, ear shape, DNA, keystroke dynamics, o
dor, signature etc. [
2
].




Page
151


I
NTERNATIONAL
J
OURNAL OF
C
OMPUTER
A
PPLICATION

I
SSUE
2,

V
OLUME
2

(APRIL

2012)





ISSN:

2250
-
1797


Figure 2
: Set of informati
on schema



A biometric system can be classified into two modules
-

(i) Database Preparation Module and (ii) Verification Module. The Database Preparation
Module consists of two sub
-
modules, and they are (a) Enroll Module and (b) Training Modu
le while the other module, Verification module can be
divided into two modules (a) Matching Module and (b) Decision Module.

1.1
Classes of biometrics
characteristics


Physiological:

related to the body of the person. The most famous unique trait is the fin
gerprint, but there are a lot more, like the shape of the hand or
of the face, the iris recognition or the DNA analysis.


Behavioral:

related to the behavior of the person. In this case the well
-
known example is the signature; another known biometric chara
cteristic
known is the voice. There is also the keystroke dynamic or the gait (the study of locomotion).


Advantages:

Biometrics has no risk of


o

Forgetting it

o

Losing it

o

Getting it stolen

o

Getting it copied

o

Fake detection

o

Being used by anyone else.


1.2
Phys
ical vs. Behavioral:



Physical



Fingerprint



Iris



Ear



Face



Retina



Hands



Behavioral



Signature



Walking gait



Typing patterns



Both



Voice


1.3: Properties of
good
B
iometrics

/

Evaluation
P
arameters

Biometric satisfy the following chara
cteristics:

1.

Universal:

Every person must possess the characteristic/attribute. The attribute must be one that is universal and seldom lost to accide
nt or
disease.



Page
152


I
NTERNATIONAL
J
OURNAL OF
C
OMPUTER
A
PPLICATION

I
SSUE
2,

V
OLUME
2

(APRIL

2012)





ISSN:

2250
-
1797

2.

Invariance of properties:

They should be constant over a long period of time. The attribute s
hould not be subject to significant differences
based on age either episodic or chronic disease.

3.

Measurability:

The properties should be suitable for capture without waiting time and must be easy to gather the attribute data passively.

4.

Singularity:

Each ex
pression of the attribute must be unique to the individual. The characteristics should have sufficient unique properties
to distinguish one person from any other. Height, weight, hair and eye color are all attributes that are unique assuming a pa
rticularly

precise
measure, but do not offer enough points of differentiation to be useful for more than categorizing.

5.

Acceptance:

The capturing should be possible in a way acceptable to a large percentage of the population. Excluded are particularly
invasive techno
logies, i.e. technologies which require a part of the human body to be taken or which (apparently) impair the human body.

6.

Reducibility:

The captured data should be capable of being reduced to a file which is easy to handle.

7.

Reliability and tamper
-
resistanc
e:

The attribute should be impractical to mask or manipulate. The process should ensure high reliability
and reproducibility.

8.

Privacy:

The process should not violate the privacy of the person.

9.

Comparable:

Should be able to reduce the attribute to a state t
hat makes it digitally comparable to others. The less probabilistic the
matching involved, the more authoritative the identification.

10.

Inimitable:

The attribute must be irreproducible by other means. The less reproducible the attribute, the more likely it w
ill be authoritative.

1.4
Performance Measurement

The performance parameter is a little bit more complicated than the others. In this case we have to define different kinds of

error or mistake that a
biometric characteristic can perform during the acquisit
ion phase or

during the recognition phase [3
].

Every biometric characteristic or parameter whether it is physiological or behavioral has to be recorded into a database and
then, when needed, the
data have to be retrieved and verified or evaluated to check
that the physical authenticity of the owner.

1.5:
Various performances metric

o

The first possibility is the FA error (False Accept): the system incorrectly declares a successful match between the input an
d a random non
-
matching pattern in the database. Thi
s error leads to calculate a FAR (False Accept Rate) that is a percent of these invalidity matches.

o

Another common rate to calculate is the FRR (False Reject Rate): in this case the inputs do not match with its corresponding
pattern.

o

The EER (Equal Error

Rate) is the error when both accept and reject rate are equal. More the EER value is then more accurate the system
is.

o

FTE or FER (Failure to Enroll) is the percent of data that fails inserting into the system (poor or invalid quality of the da
ta used as

input).

o

FTC (Failure to Capture): this error happens when, in automatic system the biometric characteristics are presented correctly
but the system
was unable to detect them or capture them correctly.

o

Template Capacity: is the maximum capacity of the sy
stem.


This small introduction about biometric procedures and measurements was useful to better comprehend the following examples of

biometrics. The
description will be deeper for some biometric (like fingerprint). Other biometrics will be described shortl
y (like iris recognition or voice analysis).



Fig
ure 3
: Biometric System Process Flow


2:
Physiological biometrics


2.
1: Fingerprint:

Currently fingerprint is the most widely used and widely accepted form of biometric technology.



Page
153


I
NTERNATIONAL
J
OURNAL OF
C
OMPUTER
A
PPLICATION

I
SSUE
2,

V
OLUME
2

(APRIL

2012)





ISSN:

2250
-
1797


Figure 4:
Fingerprint

example


Fingerprint biometric is most commonly used biometric technique, because of their uniqueness, performance and consistency ove
r the time, rich
feature information fingerprints has been used for many decades for identification and verification purp
oses. Fingerprint authentication systems are
secure, fast, reliable and easy to use. In fact the fingerprint is unique; this means that every person has his personal fin
gerprint. It is permanent,
because growing a person always keeps the same fingerprint
and the performance analysis is high: the time to analyze an input with a pattern is pretty
fast and the FRR and FAR are low.


Figure
5
: Fingerprint reader example

To clearly explain how a fingerprint reader works we need first know what kind of fingerpri
nt there are, basically there are three kinds of
fingerprints: arch, loop and whorl.




The arc fingerprints are formed by lines that enter from the left side, they form an arc in the middle of the pattern and the
n they exit to the
right side.



The loop finge
rprints are formed by lines that enter from one side of the pattern, then they form a curve and they exit to the same side.



In the whorl fingerprint there is a central point around it the lines turnaround.


The fingerprint sensor simply reads these lines a
nd creates a pattern that can be stored into a database or in another storage support. When the pattern
is created there are some algorithms that can be used to compare the input with all the entry into the database.

One of these algorithms is the Minutia
-
Based algorithm which use some minutia
-
points (like bifurcations, ridge ending and short ridge) to create
some hotspots that can be used like referring
-
point when an input is compared with the patterns into the database. In each image can be from 10 to
100

minutia points and with a match of 7
-
20 of them the pattern surely match.

Fingerprint identification is a popular personal identification method for several key reasons:



Fingerprints do not change over time.



All fingerprints are unique; even identical twi
ns have different sets of fingerprints.



Fast enrollment and matching of fingerprint; easy to use.



Low
-
cost implementation



Unique identification is accepted worldwide.



Able to store up to ten fingers for each personal enrollment in case of potential injury
to the hand.

Advantages



Very little time is required for enrolment with a fingerprint scanning system



As noted previously, fingerprints are unique identifier specific to the individual



As most people are familiar with the use of fingerprint for identificat
ion purposes, it is generally accepted as a technology

Disadvantages



Page
154


I
NTERNATIONAL
J
OURNAL OF
C
OMPUTER
A
PPLICATION

I
SSUE
2,

V
OLUME
2

(APRIL

2012)





ISSN:

2250
-
1797



Injury, whether temporary or permanent, can interfere with the scanning process



There is nothing to suggest that the same technology that is used to store fingerprint as statistical algor
ithm cannot also be used or
modified to recreate accurate depiction of the print itself



User acceptance, fingerprint scanning crosses the fine line between the impersonal and non
-
intrusive nature of passwords and PIN


2
.2
:
Face Recognition

The biometric fa
ce recognition becomes more and more popular and used. In fact it is not intrusive i.e. they do not need to touch anything du
ring the
acquisition stage to, it is fast but the main weakness is that about 10% of cases are False Rejected. Also face is primary

focus of attention in social
intercourse and important in conveying identity and emotion.

Its principal robustness in face biometric is the fast collectability. It is easy to collect data and store them into the dat
abase, for this reason the people
accept

that technology more than an intrusive one (like the DNA scan or the iris scan). In the 2000 presidential election the Mexica
n government
used this system to prevent the multi
-
voting. Some individual were recognized, they tried to vote twice using differe
nt names.

The principal weakness is that uniqueness of individuals is not so precise; two people can be alike and the input can be acce
pted even if there is not a
pattern of that specific person.

The face recognition

algorithm (face biometric) is similar to the fingerprint. It uses some unique landmarks like the distance between the eyes, t
he
depth of the eyes socket, and the width of the nose, the length of the jaw line and the s
hape of the cheekbones (Figure 6
). The
se nodal points are
measured creating a numerical code, called face print, representing the face in the database. When an input is inserted to be

checked the program
create a new face print number and check if into the database there is another similar num
ber. There is a threshold that can be modified: more this
value is high, more the program will generate False Repentance.





Figure
6
: Face recognition example



In general, face recognition techniques can be divided into two groups based on the face representation they use.


1) Appearance
-
base
d which uses a holistic texture features and is applied to either whole face or specific regions in a face image.

2) Feature
-
based which use geometrical facial features (mouth, eyes, cheeks, etc.) and geometric relationships between them.

Advantages



Easy t
o use and requires no special training or equipment.



Relatively inexpensive hardware compared to other biometrics.



Completely non
-
intrusive for user to use.

Disadvantages



It cannot be used to distinguish identical twins.



It can have high False Acceptance R
atio for people with similar features such as family members and relatives.



A person's facial features can change over time.



Environmental lighting condition might influence the recognition process.




Page
155


I
NTERNATIONAL
J
OURNAL OF
C
OMPUTER
A
PPLICATION

I
SSUE
2,

V
OLUME
2

(APRIL

2012)





ISSN:

2250
-
1797

2.
3: Iris Recognition

The Iris is muscle within the eye t
hat regulates the size of the pupil, thus controlling the amount of light entering the eye. The automated method of
iris recognition is relatively new and rapidly expanding method of biometric authentication. Iris recognition method of biome
tric authentica
tion uses
pattern recognition techniques based on high resolution images of the
Iris

of an individual’s eyes. The technology works well in both verification and
identification modes. Current systems can be used even in the presence of eyeglasses and contac
t lenses. It does not require physical contact with a
scanner. The technology is not intrusive.



Figure
7
: Iris target example


One of the more robust characteristic of the iris is its uniqueness; every person has his own iris and it is impossible that
t
wo people has the same even
the twins do not have the same iris as well. Another good aspect of the iris is the permanence; everyone has always the same
iris. The performance is
another good aspect; the algorithm is fast and the FFR and FAR are very low. T
he iris is an internal organ, so it is protected against damage (more
than fingerprints).

One disadvantage is that if a person does not want to cooperate the iris reading procedure becomes very difficult or that the

image quality influences
the FRR.

Advant
ages



It is an internal organ that is well protected against damage and wear by a highly transparent and sensitive membrane



Highly accurate biometric identification,

the fine texture remains remarkably stable over many decades
.



The iris has a fine texture t
hat


like fingerprint


is during embryonic gestation



The iris patterns are extremely complex and difficult to duplicate.



Very low False Acceptance Rate and False Rejection Rate.

Disadvantages



Expensive implementation as compared to fingerprint and facial

recognition.



It is a difficult technology if the distance is larger than a few meters and if the person is not standing still i.e.
requires user's eye to be in
close proximity to the sensor



As with other photographic biometric technologies, iris recogniti
on is susceptible to poor image quality, with associated failure to enroll
rates.


3:
Behavioral biometrics

3.
1:
Signature

Signature is Behavioral biometric and commonly used in government, commercial and legal transactions. The technology is based

on meas
uring the
signing speed, pressure applied, angle of stroke, size of signature used by the person when a signature is produced. This tec
hnology uses the dynamic
analysis of a signature to authenticate a person. One focus for this technology has been e
-
busin
ess applications and other applications where signature
is an accepted method of personal authentication.



Page
156


I
NTERNATIONAL
J
OURNAL OF
C
OMPUTER
A
PPLICATION

I
SSUE
2,

V
OLUME
2

(APRIL

2012)





ISSN:

2250
-
1797


Figure 8
: Signature example

It is easy to copy the image of the signature but nearly impossible to mimic the behavior of signing. The positive chara
cteristics of signature
biometric are the collectability and acceptability; everyone can have a personal signature in less than one second, because i
t is enough to put a sign
over a paper to confirm two fundamental things:



The provenience of the document.



The intention of an individual with regard to that document.


The signature has also some weakness in a biometric domain, first of all the uniqueness; someone could have the same signatur
e of another. The
permanence is another bad aspect for signature; unt
il a person does not learn to write cannot have a signature, and the signature could change
growing; the signature can also being copied very easily. It is clear that the signature cannot be used as a provable biometr
ic test to recognize a
person.

Advantag
es



Natural and intuitive; the technology is easy to be explained and trusted.



When combined with the pen speed, timing, and pressure, it is very difficult to imitate.



Widely accepted form of identification throughout our history.

Disadvantages



Requires mor
e expensive digitizing tablet for capturing pen pressure.



There are some inconsistencies to a person's signature.

3.
2: Voice

Voice recognition (speaker recognition) is a process of identifying people form their voice. The user

just has his voice as input a
nd the pattern of his
voice is verified with the database.
Voice recognition is emerging best candidate for future biometric security system.


Figure 9
: Voice pattern example

One of the weak
points

in voice biometric is the uniqueness, a voice can easily
being copied by another person. In addition the voice is not permanent
parameter, a person growing can change his voice and his tone can become completely different from the voice previously recor
ded to identify him.
This aspect also affects the performanc
e parameters.

Voice recognition can be used over long distances; the technology needs little additional hardware by using existing micropho
nes and voice
-

transmission technology allowing recognition over long distances via ordinary telephones.



Page
157


I
NTERNATIONAL
J
OURNAL OF
C
OMPUTER
A
PPLICATION

I
SSUE
2,

V
OLUME
2

(APRIL

2012)





ISSN:

2250
-
1797

Advantages



E
asy to use and requires no special training or equipment.



Reliable

and relatively inexpensive compared to other biometrics.



Difficult to forge.



Produce a detectable audit trail.



Consumers prefer to use voiceprints over other biometric technology for identi
fication according to a Chase bank’s research study.


Disadvantages



When processing a person’s voice over multiple channels such a microphone and then over a telephone reduces the recognition r
ate.



Physical conditions of the voice, such as those due to sic
kness, affect the voice verification process.



Environment noise reduces the overall accuracy and effectiveness of the recognition.



The storage requirement for voiceprint database can be very large.



A person’s voice changes over time.


3: Other biometric ex
amples

Some biometric recognition system are not commonly used and have to be at least mentioned because either they too expensive t
o be used or in that
way the research is spending a lot of time in lasts years. The following examples are sorted by uniquen
ess, because the uniqueness is one of the more
important characteristic that a biometric should have.

3.1
High uniqueness biometric

3.1.1:
Retinal Scan

Along with iris recognition technology, retina scan is accurate and reliable biometric technology. It is

also among the most difficult to use and copy,
hence requires well
-
trained, but highly intrusive. The users have to be cooperative and patient to achieve a proper performance, it typically req
uires
the user to look into a receptacle and focus on a given p
oint for the user's retina to be scanned. Basically the retina, a thin nerve on the back of the
eye, is the part of the eye which senses light and transmits impulses through the optic nerve to the brain. Blood vessels use
d for biometric
identification are
located along the neural retina which is the outermost of the retina's four cell layers.


Figure 1
0
: Retina Scan Image

Advantages



Highly accurate biometric identification.



The retina patterns are unique

and difficult to duplicate.



Very low False Acceptance Rate and False Rejection Rate.




Page
158


I
NTERNATIONAL
J
OURNAL OF
C
OMPUTER
A
PPLICATION

I
SSUE
2,

V
OLUME
2

(APRIL

2012)





ISSN:

2250
-
1797


Figure 11
: Internal Structure of Retina

Disadvantages



Requires close contact with the user's eye; therefore, very intrusive to the user.



Requires users to remove glasses prior to scanning.



Require skill
ed

perso
n to scan the image of iris.


3.1.2:
Facial Thermograph


This biometric measure how the face dissipates the heat. Facial thermograph makes use of an infrared camera to capture the em
ission of heat
patterns that are generated by the vascular system of the f
ace. Heat that passes through facial tissue of a human being produces a unique and
repeatable pattern (aura). The captured aura is converted into data and then compared to stored auras of authorized individua
ls, at which point
possible matches are generate

along with probability percentages. The facial print does not change over time and is accurate than facial geometry
identification technologies. The facial thermograph has the same problem of retinal scan, growing the structure change.


Image
1
2
: Depict
ion of Facial Thermography Pattern Biometric

3.1.3: Odor

Body odor recognition is a contactless physical biometric that attempts to
confirm

a person’s identity by analyzing the olfactory properties of the
human body scent. The sensors that they have develo
ped are capable of capturing the body scent from non
-
intrusive body parts, such as the hand.
Each chemical of the human scent is extracted by the biometric system and converted into a unique data string.

Odor recognition is realized by the
electronic noses

(ENoses). This technique is still under development. The only positive aspect is that the permanence is very high, but becaus
e of the
weirdness the acceptance is low. Two other negative aspects are the collectability (difficult) and the performance (low
)



Page
159


I
NTERNATIONAL
J
OURNAL OF
C
OMPUTER
A
PPLICATION

I
SSUE
2,

V
OLUME
2

(APRIL

2012)





ISSN:

2250
-
1797


Figure 13
:
Schematic Diagram of ENose
[
5
]

ENose is represented as a c
ombination of two components [4
]: sensing system and pattern recognition system. The schematic representation of

ENose can be found in Figure 13
.


People with differing immunity genes p
roduce different body odors. Each human has unique body odor that is a combination approximately thirty
different odorants. The humans’ body odor recognition is still under construction the odor recognition technique is quite us
eful in real life
applicati
on.


3.1.4: DNA


DNA is an abbreviation of deoxyribonucleic acid. DNA is a unique and measurable human characteristic that is accepted by soci
ety as absolute
evidence of one’s identity. In reality DNA identification is not absolute but it has come to be co
nsidered as the best method of confirming someone’s
identity with a near perfect probability of 99.999% accuracy.



Figure:

14

DNA structure

Only identical twins share the same DNA. It does not change during the life.
DNA can be extracte
d from blood, sk
in, and hair. [6
]

The chemical structure of everyone's DNA is the same. The only difference between people (or any animal) is the order of the
base pairs, which there
are many millions of base pairs in each person's DNA. Using these sequences, every person

can be identified based on the sequence of their base
pairs.


DNA differs from standard biometrics in several ways




DNA requires a tangible physical sample as opposed to an impression, image, or recording.



DNA matching is not done in real
-
time, and curren
tly not all stages of comparison are automated.



DNA matching does not employ templates or feature extraction, but rather represents the comparison of actual samples.



Comparatively expensive biometric technique.



3.2
Medium uniqueness biometric

3.2.1 Hand
Geometry


Hand recognition uses the geometric shape of the hand for authenticating a user's identity. An individual hand features are n
ot descriptive enough for
identification. However, it is possible to devise a method by combining various individual feat
ures to attain robust verification. The only good aspect
of this biometric is that the collectability is very easily done. This biometric does not have very negative aspects, but is
not enough to define this as a
very good biometric. The biometric is essen
tially based on the fact that every individual's hand is shaped differently than another and over the course
of time the shape of the person's hand does not significantly change.



Page
160


I
NTERNATIONAL
J
OURNAL OF
C
OMPUTER
A
PPLICATION

I
SSUE
2,

V
OLUME
2

(APRIL

2012)





ISSN:

2250
-
1797


Figure 15
: Hand Geometry


Unlike fingerprint i
maging systems, hand geometry readers are not affected by natural and environmental surface details, such as lines, scars, di
rt,
and fingernails. The basic operating principle is to measure and/or record the physical geometric characteristics of an indiv
id
ual's hand. There are
numerous hand geometry scanning devices in existence and them all (currently) fall into one of two detection categories, mech
anical or image
-
edge
detection.

Advantages



Easy for end user to use.



Hand geometry data is easy to collect.



T
his form of biometric capture is non
-
invasive to the user.


Disadvantages



The human hand geometry isn't quite as unique as other biometrics such as fingerprint.



The False Acceptance Rate can be high compared to other techniques.

3.2.2:
Hand Veins

This biom
etric is similar to hand geometric, but the collectability is a little bit more difficult to perform (Figure
16
). The system identifies a person
using the patterns of veins in the back of the hand, face, or for that matter any body part with visible veins.

A persons vein patterns are in fact highly
stable throughout their life. They are developed before birth and even differ between twins of all types. Vascular pattern re
cognition technology has
been developed to minimize the disadvantages of commercially a
vailable biometric systems and to provide users with impeccable security, usability,
reliability, accuracy, and user acquiescence.


Figure
1
6
: Hand veins example
.



Page
161


I
NTERNATIONAL
J
OURNAL OF
C
OMPUTER
A
PPLICATION

I
SSUE
2,

V
OLUME
2

(APRIL

2012)





ISSN:

2250
-
1797

3.2.3:
Ear Canal


The shape of the outer ear, lobes, bone structure and the size are unique
to each person. Ear pattern recognition is employed as a physical contactless
biometric and

uses an Optophone to verify the shape of the ear. A French company, ART Techniques, developed the Optophone and the process.
It is
a telephone type handset, which i
s comprised of two components (lighting source and cameras).

Much like the minutiae points of a palm print or fingerprint the outer ear has many detailed features that can be measured an
d compared to a
biometric template.


Figure
17
:

Parts of Human Ear
.

3.3
Low uniqueness biometric

3.3.1:
Keystrokes Dynamics

Keystroke dynamics is the process of analyzing the way a user types on a keyboard and identify him based o
n his habitual typing rhythm [5
].

Keystroke
is the
biometric

solution acquired over a certain time period the user is typing on his or her keyboard
. Keystroke
monitors and analyses all
keyboard
be
havior

performed by the user during his/her access. Based on this keystroke
behavior

performed in comparison to the user’s normal
behavior

access is granted when this user is also authorized
. Keystroke dynamics is a behavioral biometric. Keystroke dynamic
s is not what you
type, but how you type


Figure

1
8
:
How you type might protect and/or identify you
.

Advantages



Non
-
intrusive and wide user acceptance.



Natural authentication mechanism for computer and network security.



Continuous verification (monitoring
) is possible.



Minimal training and no additional hardware is required.



The user behavior becomes the identification token.

Disadvantages



High false reject rate (FAR).



Sensitive to changes in keyboard, user’s physical condition (fatigue or illness) and oth
er operational conditions.



Narrow range of applications.



Need to account for problems like typing errors.

3.3.2:
Gait

A particular way or manner of moving on foot is a
definition for gait [7
]. This can be used as a Biometric parameter because very person h
as his or
her own way of walking. From early medical studies it appears that there are twenty
-
four different components to human gait, and that if all the


Page
162


I
NTERNATIONAL
J
OURNAL OF
C
OMPUTER
A
PPLICATION

I
SSUE
2,

V
OLUME
2

(APRIL

2012)





ISSN:

2250
-
1797

measurements are considered, gait is unique. This has made gait recognition an interesting topic to b
e used for identifying individuals by the manner
in which they walk.

Figure 1
9

illustrates the complex biological process of the musculo
-
skeletal system, which can be divided into several types of
sub events of human
-
gait. The instances that are shown in
this figure are used to extract parameters for being used as an identification system of each
individual.


The analysis of biometric gait recognition has been studied for a longer period of time for the use in identification, survei
llance and forensic syst
ems
and is becoming important, since it can provide more reliable and efficient m
eans of identity verification [8
].


Figure
19
: Division of the gait cycle into five phase periods and two swing phase periods
.

4:
Multimodal Biometric Systems

Multimodal biom
etric systems are those that utilize more than one physiological or behavioral characteristic for enrollment, verification, o
r
identification [4]. In applications such as border entry/exit, access control, civil identification, and network security, mu
lti
-
modal biometric systems
are looked to as a means of

1.

Reducing false non
-
match and false match rates,

2.

Providing a secondary means of enrollment, verification, and identification if sufficient data cannot be acquired from a give
n biometric
sample, and

3.

Combati
ng attempts to fool biometric systems through fraudulent data sources such as fake fingers.

A multimodal biometric verification system can be considered as a classical information fusion problem i.e. can be thought to

combine evidence
provided by different

biometrics to improve the overall decision accuracy. Generally, multiple evidences can be integrated at one of the following
three levels.



Abstract level:
The output from each module is only a set of possible labels without any confidence value associated

with the labels; in
this case a simple majority rule may be used to reach a more reliable decision.



Rank level:

The output from each module is a set of possible labels ranked by decreasing confidence values, but the confidence values
themselves are not sp
ecified.



Measurement level:

the output from each module is a set of possible labels with associated confidence values; in this case, more
a
ccurate
decisions can be made by integrating different confidence values.

5: C
hoice of Biometric

There are several Bi
ometrics techniques available as discussed above. Each technique has its advantages and disadvantages. A brief comparison of
different techniques is given below in the Table 1
.




6: Conclusion

People acceptance in biometrics depends on different factors,

from the society where they live to the personal perception of what is considered. The
introduction of biometrics techniques in a too fast and uncontrolled w
ay could lead to a reject of the
se technologies. There are also differences
between biometrics, pe
ople are less threatened to use voice or facial recognition, rather than iris and fingerprint, because the human normal behav
ior
is to identify other persons by the face and the voice and other behavioral and physical details that are saw by everyone. Th
e
risk of all biometrics is
that there is a possibility that can be forged.

Another important subject treated is the storage of biometrics data.

In conclusion the importance of


Page
163


I
NTERNATIONAL
J
OURNAL OF
C
OMPUTER
A
PPLICATION

I
SSUE
2,

V
OLUME
2

(APRIL

2012)





ISSN:

2250
-
1797

biometrics will certain grow during next years and will interest all kind of fiel
d, from economical to juridical and ethical, involving industry,
governments and academic sectors, in order to create and improve security standards.

Table 1: Comparison of various biometrics

*M=
Medium
, *H=
High
, *L=
Low

































References


[1] Fernando L. Podio and Jeffrey “Biometric Authentication Technology”.

http://www.copacommission.org/meetings/hearing1/dunn
-
podio.test.pdf
.


[2]
http://www.cse.iitk.ac.
in/users/biometrics /ages/what_is_biom_more.html
.


[3
]
http://diuf.unifr.ch/main/is/sites/diuf.unifr.ch.main.is/files/file/
course
s/eBiz_fs08/ imasoni_ Cavadini_
Introducing
_ePass.pdf
.


[4] P. Keller,

"
Overview of Electronic Nose Algorithms
",
International

Joint Conference of Neural Networks (IJCNN'99)
,
Washington
, USA, 1999.


[5] Anil K. Jain , Biometric Authentication based on Keystroke Dynamics
http:// www.c
se.msu.edu/~cse891/Sect601/ KeystrokeRcg.pdf
.


[6
]
http://www.cse.msu.edu/~cse891/Sect 601CaseStudy/DNABiometricIdentifier.pdf
.


[7
]
http://biometrics.derawi.com/?Page _id=38


[8]
Biometric Gait Recognition,
Jeffrey E. Boyd and James J. Little.


Universality

Uniqueness

Permanence

Collectability

Performance

Acceptability

Circumvention

Fingerprin
t

M

H

H

M

H

M

H

Face

H

L

M

H

M

M

M

Iris

H

H

H

M

H

L

L

Signature

L

L

L

H

L

H

H

Voice

M

L

L

H

L

M

H

Retina

H

H

M

L

H

L

L

Facial Thermograph

H

H

L

H

M

H

H

Odor

L


L


L

M


L

M


M

Hand Geometry

M


M


M


H


M


M


M


DNA

H

H


H

L

H

L


L

Hand Vein

M

M

M


M


M


M


L


Ear

H


M


H


M


M


H


M


Keystroke

L

L

L

M

L

M

M

Gait

M

L

L

H

L

H

M