Chapter 1 individual.

For a layman, it could be said

that biometrics are the science of measuring

1. INTRODUCTION: physical and/or behavioural characteristics

that are unique to each individual and they
verify that an individual is who he or she
claims to be.

Often seen in science-fiction action-

adventure movies, face pattern matchers
and body scanners seem about to emerge as
replacements for computer passwords.
Thus, biometric techniques are of interest in
any area where it is important to verify the
true identity of an individual. Initially, these

The terms "Biometrics" and "Biometry" techniques were employed primarily in

have been used since early in the 20th specialist high security applications,

century to refer to the field of development however we are now seeing their use and

of statistical and mathematical methods proposed use in a much broader range of

applicable to data analysis problems in the public facing situations.

biological sciences. Recently, these terms

Fingerprint, facial, or other biometric
have also been used to refer to the emerging
data can also be placed on a smart card and
field of information technology devoted to
users can present both the smart card and
automated identification of individuals
their fingerprints or faces to merchants,
using biological traits, such as those based
banks, or telephones for an extra degree of
on retinal or iris scanning, voice patterns,
authentication.
dynamic signatures, fingerprints, face
recognition, or hand measurements, 1.1 History of Biometrics:
especially for authentication purposes.
It is tempting to think of biometrics as
Thus, biometrics can be defined as the
being sci-fi futuristic technology that we
science and technology of measuring and
shall all be using some time in the near
statistically analysing biological data. They
future. But actually, the basic principles of
are measurable physiological and / or
biometric verification were understood and
behavioural characteristics that can be
practiced somewhat earlier. Thousands of
utilized to verify the identity of an

1
years earlier, the people in the Nile valley respectable numbers of devices and poised
routinely employed biometric verification for significant growth as large-scale
formally identify individuals via unique applications have started to unfold.
physiological parameters such as scars and
a combination of features such as 1.2 Basic Techniques:
complexion, eye colour, height and so on.
Though they did not use the advanced A biometric system is a real-time

technological tools that we do today, the identification system which identifies a

basic principles used by them were similar. person by measuring a particular physical
or behavioural characteristic and later
Later, in the nineteenth century there comparing it to a library of characteristics
was a peak of interest as researchers into belonging to many people. Fingerprint and
criminology attempted to relate physical other biometric devices consist of a reader
features and characteristics with criminal or scanning device, software that converts
tendencies. This resulted in a variety of the scanned information into digital form,
measuring devices being produced and and wherever the data is to be analysed, a
much data being collected. The results were database that stores the biometric data for
not conclusive but the idea of measuring comparison with previous records. When
individual physical characteristics seemed converting the biometric input, the software
to stick and the parallel development of identifies specific points of data as match
fingerprinting became the international points. The match points are processed
methodology among police forces for using an algorithm into a value that can be
identity verification. compared with biometric data scanned
when a user tries to gain access.
But, for many years, a fascination with
the possibility of using electronics and the
Thus, biometric devices can be
power of microprocessors to automate
explained with a 3-step procedure. They
identity verification had occupied the
are,
minds of individuals and organizations both
in the military and commercial sectors. 1. A sensor takes an observation. The type
Hence, various projects were initiated to of sensor and its observation depend on the
look at the potential of biometrics. In the type of biometrics device used. This
last decade, the biometric industry has observation gives us a Biometric Signature
matured to a global industry shipping of the individual.

2
2. A computer algorithm normalizes the Identification. Identification means
biometric signature so that it is in the same establishing a person's identity Verification
format (size, resolution, etc.) as the involves confirming or denying a person's
signatures on the system's database. The claimed identity. Each one of these
normalization of the biometric signature approaches has its own complexities and
gives us an A Normalized Signature of the could probably be solved best by a certain
individual. biometric system. Verification systems are

3. A matcher compares the normalized generally described as a 1-to-1 matching

signature with the set (or sub-set) of system because the system tries to match

normalized signatures on the system's the biometric presented by the individual

database and provides a similarity score against a specific biometric already on file.

that compares the individual's normalized Identification systems are different from

signature with each signature in the verification systems because an

database set (or sub-set). What is then done identification system seeks to identify an

with the similarity scores depends on the unknown person, or unknown biometric.

biometric system's application. The system tries to answer the questions

“Who is this person?” or “Who generated
Thus, a biometric system is this biometric?” and must check the
essentially a pattern recognition system, biometric presented against all others
which makes a personal identification by already in the database. Identification
determining the authenticity of a specific systems are described as a 1-to-n matching
physiological, or behavioural characteristic system, where n is the total number of
possessed by the user. An important issue biometrics in the database. Forensic
in designing a practical system is to databases, where a government tries to
determine how an individual is identified. identify a latent print or DNA discarded at
Depending on the context, a biometric a crime scene, often operate as
system can be either a verification identification systems.
(authentication) system or an identification
system.

1.2.1 Verification Vs Identification:

There are two different ways to

resolve a person's identity: Verification and
Fig. 1.1: Identification and Verification

3
 Chapter 2 large user bases. Fingerprint verification
may be a good choice for in house systems
where adequate explanation and training
2. BIOMETRICS METHODOLOGIES:
can be provided to users and where the
Fingerprint recognition and iris
system is operated within a controlled
scanning are the most well-known forms of
environment. It is not surprising that the
biometric security. However, facial
workstation access application area seems
recognition and (finger and palm) vein
to be based almost exclusively around
pattern recognition is also gaining in
fingerprints, due to the relatively low cost,
popularity.
small size (easily integrated into
Fingerprint verification – keyboards) and ease of integration.

 Hand geometry -

Fig 2.1: Fingerprint Verification

There are a variety of approaches to

fingerprint verification. Some of them try to
emulate the traditional police method of
Fig 2.2: Hand Geometry
matching minutiae, others are straight
pattern matching devices, and some adopt a Hand geometry is concerned with
unique approach all of their own, including measuring the physical characteristics of
ultra-sonics. There are a greater variety of the users hand and fingers, from a three-
fingerprint devices available than other dimensional perspective. One of the most
biometric systems at present. established methodologies, it offers a good
balance of performance characteristics and
Potentially capable of good accuracy
is relatively easy to use. This methodology
(low instances of false acceptance)
may be suitable where we have larger user
fingerprint devices can also suffer from
bases or users who may access the system
usage errors among insufficiently
infrequently and may therefore be less
disciplined users (higher instances of false
disciplined in their approach to the system.
rejection) such as might be the case with

4
Accuracy can be very high if desired. Hand biometrics leading to the perception of
geometry readers are deployed in a wide voice verification as unfriendly in some
range of scenarios, including time and quarters. However, much work has been
attendance recording where they have and continues to be undertaken in this
proved extremely popular. Ease of context and it will be interesting to monitor
integration into other systems and progress accordingly.
processes, coupled to ease of use makes
hand geometry an obvious first step for  Retinal scanning -

many biometric projects.

 Voice verification -

Fig 2.4: Retinal Scanning

Fig 2.3: Voice Verification This is an established technology where

the unique patterns of the retina are scanned
This is a potentially interesting by a low intensity light source via an optical
technique if the amount of voice coupler. Retinal scanning has proved to be
communication that takes place with regard quite accurate in use but does require the
to everyday business transactions is user to look into a receptacle and focus on
considered. Some designs have a given point. This is not particularly
concentrated on wall mounted readers convenient for spectacle wearers and for
whilst others have sought to integrate voice those who avoid intimate contact with the
verification into conventional telephone source used for the scan and hence this has
handsets. Whilst there have been a number a few user acceptance problems although
of voice verification products introduced to the technology itself can work well. It is
the market, many of them have suffered in believed to replace traditional ID methods
practice due to the variability of both such as P.I.N. numbers for accessing
transducers and local acoustics. In addition, A.T.M.s and virtually every other
the enrolment procedure has often been electronic device used for conducting
more complicated than with other

5
business where identification is a Signature verification enjoys a synergy
requirement and prerequisite. with existing processes that other

 Iris scanning - biometrics do not as people are used to

signatures as a means of transaction related
identity verification and mostly see nothing
unusual in extending this to encompass
biometrics. Signature verification devices
have proved to be reasonably accurate in
operation and obviously lend themselves to
applications where the signature is an
Fig 2.5: Iris Scanning accepted identifier.

Iris scanning is the less intrusive of the  Facial recognition -

eye related biometrics. It utilizes a
conventional camera element and requires
no intimate contact between user and
reader. In also has the potential for higher
than average template matching
performance. It has been demonstrated to
work with spectacles in place and with a
variety of ethnic groups and is one of the Fig 2.7: Facial Recognition

few devices that can work well in

Facial recognition devices have been
identification mode. However, ease of use
difficult to substantiate in practice and
and system integration have not
extravagant claims have sometimes been
traditionally been strong points with the iris
made them. Facial recognition is very
scanning devices.
attractive from the user perspective and
 Signature verification -
they may eventually become a primary
biometric methodology.

Fig 2.6: Signature Verification

6
 Chapter 3

3. BIOMETRICS DESIGN:

The design of the system consists of

the following important parameters

1. Scanning-using DSP processor.

2. Searching-based on the principal of Fig 3.2: Process Diagram

GOOGLE SEARCH.
The detailed description of each and
3. Networking-all the election booths are every internal unit in the VOTING
connected in a network. SYSTEM is given below. It can be divided
in to the following main categories.
4. Data transfer-using telephone lines.

The only pre-requisite for the use of

this finger print scanner is a personal
identification card. We hope that this
system proves to be efficient and enables
the people to be smarter in choosing their
leaders.
3.1 Description of The Voting Machine:

Fig 3.3: Voting System

Finger print scanner:

The finger print scanner consists of

the following parts:

1. Finger Print Sensor: This is used to

scan the thumb impression. The data
obtained from this is analogue in nature.
This is transferred to the A/D converter for
Fig 3.1: Block diagram for the whole further processing.
process in brief

7
2. A/D Converter: This is used to convert any one of the finger prints stored in the
the analogue data from the SENSOR into data base then the vote is rejected.
the digital data and transfer it to the
processor.
3. Flash Rom: This is for the storage of the
data temporarily in the DSP processor. This
will work until the data is transferred to the
main memory of the HOST.
4. DSP Chip: This is used to receive the Fig 3.5: Finger Print Expression

data and process it. This is connected to the Feature extraction and comparison
USB port for further transfer of the data. Scanning and processing:
5. USB Port: The sole purpose of the USB
The biometric sensor scans the
port is to establish a communication link
image. This so scanned data is in the
between the DSP processor and the
analogue form. This is converted into
MEMORY (DATABASE).
digital form by using an A/D converter.
Extraction of thumb impression:
Since the image is to be transferred quickly
to the converter, it is interfaced in parallel
with the DSP chip. The data received from
the parallel in ports is further processed in
the chip. Parallel interfacing is done to have
a quick performance.

Transfer of the processed data to hard

Fig 3.4: Thumb Impression disk:

The data which is processed in the

The next step in the process is the
DSP chip (finger print)is transferred in
extraction of the thumb impression from the
parallel to the HARD DISK for searching
memory. The features of the fingerprint are
process. The BIOS language of the HARD
stored in the form of pixels. This is further
DISK is stored in SDRAM which is also
sent for pattern matching where the finger
interfaced in parallel with the chip. This
print is then compared with the records in
helps the chip to transfer the image to the
the database. If the pattern matches with
HARD DISK for further process. The
any one of the records then the vote is
image transferred to the HARD DISK is
accepted. If the feature doesn’t match with
compared with that of the DATA BASE.

8
The circuit connections for the process Here the identity of a person is checked
explained are as shown below. A DE- whether he\she belongs to the particular
MULTIPLEXER is connected to the inputs constituency or not. The machine which
in parallel to the chip. The option which is takes the finger print of the person checks it
given is transferred to the chip in binary with the data base already stored in it. If the
form. E.g.:1. Party one, 2. Party two, 3. finger print matches it will give access to
Party three, 4. Party four. the person to cast his vote and if it doesn’t
match any of the finger prints stored in the
Suppose if the option is two then the data base then it will reject the voter. Thus,
binary data transfer to the ports is 0010.DE- his method will enable the members of that
MULTIPLEXER is designed according to particular constituency only to vote. This
the number of parties present in the can be taken as the first step to avoid
constituency. The transferred data is stored rigging. To have a faster performance the
in a 16-bit register according to their option. searching technique is implemented on the
After counting takes place the count of each basis of GOOGLE SEARCH. The process
party is transferred to the transmitter. A in the form of a flow chart is as given
transmitter is interfaced serially to the DSP below.
chip. The transmitter data is transferred
through the telephone lines to the HOST. Processing and counting:
The counts of similar parties are added at
the host and the data is transferred to the
STATE HEAD QUARTERS to display the
result of the particular constituency.

Block diagrams for various processes Fig 3.7: Counting

verification of voters: After the person gets the

PROCEED symbol from the voting
machine, there appears a screen on which
all the party’s names along with the
symbols are present. The person should
Fig 3.6: Verification of Votes select any one of the party by giving the
number allotted to that particular party as
input through the keypad. After the option

9
is selected the voter is prompted for a searched folder. The searched folder is
confirmation. In case the voter enters an programmed such that an image cannot be
invalid number, the screen reappears and present more than once in this folder. So,
he\she is prompted to cast the vote again. when a voter casts multiple votes the
Then according to the option selected, the exception is generated and an alarm is
vote count of the particular party gets raised and even the police can be informed
incremented by one. Finally, at the end of about the identity of the intruder indulging
the day, the position of the parties in terms in this illegal activity. This is shown in the
of the total votes cast can be known. A very block diagram given below the scanned
simple flow chart for the above process is vote is first checked with the acceptability
as shown below: of the voter as explained in the first flow
chart. If the finger print is accessible then
Rejection of voter: the data of the specified person is taken into
account. The voter’s thumb impression is
verified with the previously cast votes. If
there is no match then the vote is accepted
and the count is increased by one. If the
vote matches with any of the previous votes
then the vote is rejected and the person’s
Fig 3.8: Rejection of Voter
identity is stored and it is given to the police
After we have emerged out with a for further enquiry. There is a flash ROM in
solution to check voter’s identity in a which these details can be stored.
constituency, our next task is to see that a
particular person cannot vote more than 3.2 ADVANTAGES AND
once i.e., to avoid multiple votes. This task
DISADVANTAGES
can be accomplished by simple software
technique employed. It consists of two A. Advantages: -

folders namely searched and unsearched. • Cost effective

Initially the searched folder consists of no • This system allows only authenticated
images. The thumb impression images of voting than the existing equipment as
all the voters of a constituency will be the person is identified based on his
present in the unsearched folder. As and Fingerprint which is unique to each
when a vote is cast, the image of the individual.
particular voter gets transferred to the

10
 • Low power consumption 3. It could also be used to conduct
general assembly elections where
• It is economical
number of candidates are less than or
• Less manpower required
equal to eight in the current situation,
• Time conscious, less time required for on a small scale basis.
voting &counting

• Avoids invalid voting as it prevents Chapter 4

unregistered voters from voting.

• Ease of transportation due to its 4. FUTURE SCOPE:

compact size.
1. Number of active candidates may
• Convenient on the part of voter. increase.

B. Disadvantages: - 2. It could be interfaced with the

• Before voting the user has to enrol computer and result scan be
first. displayed quickly and easily or the
result can be stored in the central
• Sensitivity of finger print module
server whose data can be stored in the
causes sometimes Combine character
backend servers.
error.
3. It could be interfaced with the printer
so that the hardcopy of the data can
3.3 APPLICATIONS be taken.

This project can be used as a voting 4. It could make the results available at
machine to prevent rigging, during the any corner of the world in minimum
elections in the polling booths. time.

1. Fast track voting which could be used

in small scale elections, like resident
welfare association, “panchayat”
level election and other society level
elections, where results can be
instantaneous.
2. It could also be used to conduct
opinion polls during annual
shareholders meeting.

11
 Chapter 5

5. CONCLUSION:

The project “Fingerprint Based

Voting Machine” was mainly intended to
develop a fingerprint based advanced
Electronic Voting Machine (EVM) which
helps in free and fairway of conducting
elections which are basis for democratic
country like India.

Chapter 6

6. REFERENCES:

1. K. Agarwala, D. T. Shahani, and P. V.

Andresen. Report of the expert committee
for evaluation of the upgraded electronic
voting machine (EVM). Sept. 2006
2. R. Haenni, E. Dubuis,and U.Ultes-
Nitsche, “Research on e-voting
technologies”. Bern University of Applied
Science, Technical Report 5, 2008.
3. Balaji, Speech of Shri V S Sampath, CEC
for Defence Eststes Day Lecture 201

12