An-atm-with-an-eye.pdf 44x6g

ATM WITH AN EYE ABSTRACT

institutions alike from fraud and other breaches of security.

There is an urgent need for

improving

banking

security

region.

With

in

INTRODUCTION

the

advent of ATM though banking

The rise of technology in

became a lot easier it even

India has brought into force

became a lot vulnerable. The

many types of equipment that

chances of misuse of this much

aim

hyped ‘insecure’ baby product

satisfaction. ATM is one such

(ATM) are manifold due to the

machine which made money

exponential

transactions

growth

of

at

more

customer

easy

for

‘intelligent’ criminals day by

customers to bank. The other

day. ATM systems today use

side of this improvement is the

no more than an access card

enhancement of the culprit’s

and

probability

PIN

for

identity

verification. This situation is

‘unauthentic’

unfortunate since tremendous

Traditionally,

progress has been made in

handled

biometric

combination

identification

techniques,

including

finger

by

to

get

his share.

security

is

requiring

the

of

a

physical

access card and a PIN or other

printing, facial recognition, and

in order to access a

iris scanning.

customer’s model

.

invites

This

fraudulent

This paper proposes the

attempts through stolen cards,

development of a system that

badly-chosen or automatically

integrates Facial recognition

assigned PINs, cards with little

and Iris scanning technology

or no encryption schemes,

into the identity verification

employees with access to non-

process used in ATMs. The

encrypted customer

development of such a system

information and other points of

would

failure.

serve

consumers

www.pediain.com

to and

protect financial

1

ATM WITH AN EYE Our

paper

automatic

proposes

teller

security

model

combine

a

card,

machine that

physical

a PIN,

an

and

would

checking -- is unique to every person,

more

so

than

fingerprints.

access

electronic

2. LITERATURE REVIEW

facial recognition. By forcing For most of the past ten years,

the ATM to match a live image

the majority of ATMs used

of a customer’s face with an image

stored

database

that

in is

a

bank

associated

worldwide ran under IBM’s

now-defunct

OS/2.

However, IBM hasn’t issued a

with the number, the

major update to the

damage to be caused by stolen

operating system in over six

cards and PINs is effectively

years.

neutralized. Only when the PIN

Movement

in

the

banking world is now going in

matches the and the live image and stored image match

would

a



be

two directions: Windows and Linux. NCR, a leading world-

considered fully verified.

wide ATM manufacturer,

A system can examine just

recently

the eyes, or the eyes nose and

announced

an

agreement to use Windows XP

mouth, or ears, nose, mouth

Embedded in its next

and eyebrows, and so on.

generation In this paper , we will also

of

personalized

ATMs (crmdaily.com.) Windows

look into an automatic teller

XP Embedded allows

machine

OEMs to pick and choose from

providing

security the

model

customers

a

the thousands of components

cardless, -free way to

that make up Windows

get their money out of an ATM.

XP

Just step up to the camera

integrated

while your eye is scanned. The

networking and database

Professional,

including multimedia,

iris -- the colored part of the eye

the

camera

www.pediain.com

will

be

2

ATM WITH AN EYE management

functionality.

bulk of transaction processing

This makes the use of off-the-

(linux.org.) This model would

shelf facial recognition

also work well for the

code more desirable because it

proposed system if the ATMs

could easily be compiled for

processors were not powerful

the Windows XP

enough to quickly

environment

and

the

perform the facial recognition

networking and database tools

algorithms.

will already be in place.

In of the improvement of

For less powerful ATMs, KAL, a

security standards, MasterCard

software

is spearheading

development

company based in Scotland,

an effort to heighten the

provides

Kalignite

encryption used at ATMs. For

CE, which is a modification of

the past few decades, many

the Windows CE platform.

machines have used the Data

This

allows

developers

that

Encryption Standard developed

target older machines to more

by IBM in the mid

easily develop complex

1970s that uses a 56-bit key.

-interaction

systems

.

DES has been shown to be

Many financial institutions are

rather easily cracked,

relying on a third choice,

however,

Windows NT, because of its

computing hardware. In recent

stability

and

maturity

as

a

given

proper

years, a “Triple DES” scheme

platform.

has been put forth that uses

On an alternative front, the

three

largest bank in the south of

effective 168-bit key length.

Brazil, Banrisul, has

MasterCard now requires new

installed a custom version of

or relocated ATMs to use the

Linux in its set of two thousand

Triple DES scheme, and

ATMs, replacing legacy

by April, 2005, both Visa and

MS-DOS systems. The ATMs

MasterCard will require that

send

database

requests

to

such

keys,

for

an

any ATM that s

bank servers which do the

www.pediain.com

3

ATM WITH AN EYE their cards must use Triple

environment – would all be

DES. ATM manufacturers are

controlled

now developing newer

narrow limits, making hugely

models that Triple DES

robust

natively; such redesigns may

unnecessary

make them more

facial

amenable

to

also

including

within

certain

software .One

recognition

leading algorithm

class is called

snapshot cameras and facial

image template based. This

recognition software, more

method attempts to capture

so

than

regards

they to

would

be

retrofitting

in

pre-

existing machines . There

are

images into facial templates. Neural networks, among other

hundreds

proposed

global features of facial

and

of

actual

methods, are often used to construct these templates

implementations of facial

for later matching use. An

recognition technology from all

alternative method, called

manner

geometry-based, is to explicitly

of

vendors

for

all

manner of uses. However,

examine

the

individual

for the model proposed in this

features of a face and the

paper, we are interested only

geometrical

in the process of facial

between

verification – matching a live

(Gross.) What must be taken

image to a predefined image

into

relationship those

features

to a claim of identity – not in the process of

, though, are certain

facial evaluation – matching a

key factors that may change

live image to any image

across live images:

in a database. Further, the

illumination, expression, and

environmental

pose (profile.)

conditions

under which the verification

A

takes place – the lighting, the

conducted

imaging

recognition algorithms, notably

system,

the

image

profile, and the processing

www.pediain.com

study

was

recently

of

leading

one

4

ATM WITH AN EYE developed by two researchers

scream profile, or squinted

at MIT, Baback Moghaddam

eyes, dropped the recognition

and Alex Pentland, and

rates significantly. Lastly,

one

a

commercial

product

matching

profile

changes

from Identix called FaceIt. The

worked reasonably well when

MIT program is based on

the initial training image(s)

Principal Feature Analysis, an

were frontal, which allowed 70-

adaptation of template based

80% success rates for up to 45

recognition. FaceIt’s

degrees of profile

approach

uses

geometry-

change…

however,

based local feature analysis.

success

Both algorithms have to be

keeping ATM s content

initialized

by

providing

the

isn’t

70-80%

amenable

to

with the system.

locations of the eyes in the

The

database image, from which

draw, then, is to take a frontal

they can create an internal

image for the bank

representation

database, and to provide a

of

the

natural

conclusion

to

normalized face. It is this

prompt to the , verbal or

representation to which future

otherwise, to face the

live images will be compared .

camera directly when the ATM

In the study, it was found that

verification process is to begin,

both

so as to avoid the

programs

handled

changes in illumination

need to for profile

well. This is important because

changes. With this and other

ATM use occurs day and night,

accommodations,

with or without

recognition

artificial illumination. Likewise,

verification can rise above

the programs allowed general

90%. Also worth noting is that

expression changes

FaceIt’s local feature analysis

while

maintaining

success.

However,

expressions, such as a

matching extreme

rates

for

method handled variations in the test cases slightly better than the PGA system used by the MIT researchers .

www.pediain.com

5

ATM WITH AN EYE Another

paper

advantages

shows

more

analysis

using

local

program that can be used on a

in

facial

verification

feature analysis systems.

variety of platforms,

For internal representations of

including

faces,

processors, and to determine

LFA

stores

them

embedded

topographically; that is, it

behavior

maintains feature relationships

match

explicitly.

/ non-match cases.

Template

based

protocols

for

the

systems, such as PGA, do not. The advantages of LFA are

3. OUR METHODOLOGY

that analysis can be done on varying levels of object grouping,

and

that

The first and most important analysis

step of this project will be to

methods can be independent

locate a powerful

of the topography. In other

open-source facial recognition

words, a system can examine

program

just the eyes, or the eyes nose

feature analysis and that is

and mouth, or ears,

targeted at facial verification.

nose, mouth and eyebrows,

This

and so on, and that as better

compilable on multiple

analysis algorithms are

systems, including Linux and

developed, they can fit within

Windows variants, and should

the data framework provided

be customizable to the

by LFA

extent

The conclusion to be drawn for

variations in processing power

this project, then, is that facial

of the machines onto which it

verification

would be deployed.

software is currently up to the

We

task of providing high match

familiarize ourselves with the

rates for use in ATM

internal workings of the

transactions. What remains is

program so that we can learn

to find an appropriate open-

its strengths and limitations.

source local feature

Simple testing of this

www.pediain.com

that

uses

program

will

of

should

allowing

then

need

local

be

for

to

6

ATM WITH AN EYE program occur

will so

evaluate

also

that its

need we

to

could

effectiveness.

will be necessary, of course, because we will not have access to an actual ATM or its

Several

software.

sample images will be taken of

Both pieces of software will be

several individuals to be used

compiled

as test cases – one

Windows XP and a Linux

each

for

“”

images,

and

run

on

a

system. Once they are both

and several each for “live”

functioning properly, they will

images, each of which would

be tweaked as much as

vary pose, lighting conditions,

possible

and expressions.

performance (decreasing the

Once

a

chosen,

final we

program

will

develop

This

increase

is

time spent matching) and to

a

decrease memory footprint.

simple ATM black box program.

to

Following that, the black boxes

program

will

will

be

broken

into

two

server as the theoretical ATM

components – a server

with which the facial

and a client – to be used in a

recognition

software

will

two-machine

network.

The

interact. It will take in a name

client code will act as a

and , and then

interface, ing all input

look in a folder for an image

data to the server code, which

that is associated with that

will handle the calls to the

name. It will then take in an

facial

image from a separate folder

further reducing the memory

of “live” images and use the

footprint and processor

facial recognition program

load required on the client end.

to

generate

a

match

level

recognition

software,

In this sense, the thin client

between the two. Finally it will

architecture of many

use the match level to

ATMs will be emulated.

decide whether or not to allow

We will then investigate the

“access”, at which point it will

process of using the black box

terminate. All of this

program to control a

www.pediain.com

7

ATM WITH AN EYE USB camera attached to the

for in most cases. Further, a

computer to avoid the use of

positive visual match would

the folder of “live” images.

cause the live image to be

Lastly, it may be possible to

stored in the database so that

add

future transactions would have

some

sort

of

DES

encryption to the client end to

a broader base from which to

encrypt the input data and

compare if the original

decrypt the output data from

image fails to provide a match

the server – knowing that

– thereby decreasing false

this will increase the processor

negatives.

load, but better allowing us to

When a match is made with

gauge the time it

the PIN but not the images, the

takes to process.

bank could limit transactions in a manner agreed upon by the customer when the was opened, and could store the image of the for later

ATM SYSTEMS

examination by bank officials. In regards to bank employees gaining access to customer

Our ATM system would only attempt to match two (and

PINs for use in fraudulent

later, a few) discrete images,

transactions,

searching

would likewise reduce that

through

a

large

this

system

database of possible matching

threat to exposure to the low

candidates

would

limit imposed by the bank and

unnecessary.

The

be process

agreed to by the customer on

would effectively become an

visually

exercise in pattern matching,

transactions.

which

would

great

deal

not of

require time.

a

With

unverifiable

In the case of credit card use

at

ATMs,

such

a

appropriate lighting and robust

verification system would not

learning

currently be feasible without

software,

slight

variations could be ed

www.pediain.com

creating an overhaul for the

8

ATM WITH AN EYE entire

credit

card

issuing

withdrawn from an .

industry, but it is possible that

The bank s were not

positive

results

(at that time) connected by a

significant

fraud

(read: reduction)

computer network to the ATM.

achieved by this system might

Therefore, banks were at first

motivate such an overhaul.

very exclusive about who they

The

last

consideration

is

gave ATM privileges to. Giving

that consumers may be wary

them

of the privacy concerns raised

holders

by

used before ATM cards) with

maintaining

images

of

only

to

credit

card

(credit

cards

were

customers in a bank database,

good

encrypted or otherwise, due to

modern

possible hacking attempts or

authenticate

employee

using a plastic card with a

misuse.

However,

banking

records.

ATMs,

In

customers

themselves

one could argue that having

magnetic

the image compromised by a

encodes

third party would have far less



dire consequences than the

entering a numeric code



called

information

itself.

stripe,

by

the

customer's

number,

a

which

PIN

and

by

(personal

Furthermore, since nearly all

identification number), which

ATMs

in some cases may be changed

videotape

customers

engaging in transactions, it is

using the machine. Typically, if

no broad leap to realize that

the

banks already build an archive

incorrectly several times in a

of their customer images, even

row, most ATMs will retain the

if

card as a security precaution

they

are

grouped

not

necessarily

with



number

is

entered

to prevent an unauthorised

information.

from working out the PIN

HISTORY

by pure guesswork..

The first ATMs were off-line machines, was

meaning

not

www.pediain.com

money

automatically

9

ATM WITH AN EYE HARDWARE

AND

SOFTWARE

with Linux. Other platforms include RMX 86, OS/2 and Windows 98 bundled with Java.

ATMs

contain

secure

cryptoprocessors,

generally

The newest ATMs use Windows XP or Windows XP embedded.

within an IBM PC compatible host

computer

in

a

secure

RELIABILITY

enclosure. The security of the ATMs

machine relies mostly on the integrity

of

the

cryptoprocessor: software

secure

the

often

host

runs

commodity

on

a

operating

modem

over

a

via

a

dedicated

telephone line, although the move

towards

reliable, but if they do go wrong customers will be left without cash until the following morning or whenever they can

hours. Of course, not all errors

connect directly to their ATM Processor

generally

get to the bank during opening

system.In-store ATMs typically

Transaction

are

Internet

are

to

the

detriment

of

customers; there have been cases of machines giving out money without debiting the , or giving out higher

connections is under way.

value notes as a result of In moving

addition, away

ATMs

from

are

custom

incorrect

denomination

of

banknote being loaded in the

circuit boards (most of which

money cassettes. Errors that

are

can occur may be mechanical

based

on

architecture)

Intel

and

8086

into

full-

(such

as

card

transport

fledged PCs with commodity

mechanisms; keypads; hard

operating

disk failures); software (such

systems

such

as

Windows 2000 and Linux. An

as operating system; device

example of this is Banrisul, the

driver;

largest bank in the South of

communications;

Brazil, which has replaced the

down to operator error.

application); or

purely

MS-DOS operating systems in its automatic teller machines

www.pediain.com

10

ATM WITH AN EYE SECURITY

when you're alone, leave the Early ATM security

area.

focused on making the ATMs invulnerable to physical attack;

Keep you PINconfidential.

they

were

Memorize

with

dispenser

effectively

safes

mechanisms.

your

Identification

Personal

Number

(PIN);

ATMs are placed not only near

don't write it on your card or

banks, but also in locations

leave it in your wallet or purse.

such as malls, grocery stores,

Keep your number to yourself.

and

Never provide your PIN over

restaurants.

The

other

side of this improvement is the

the telephone, even if a caller

enhancement of the culprit’s

identifies himself as a bank

probability

employee or police officer.

to

get

his

‘unauthentic’ share.

Neither person would call you

ATMs are a quick and

to obtain your number.

convenient way to get cash. They

are

also

public

and

Conduct

transactions

in

visible, so it pays to be careful

private. Stay squarely in front

when

of the ATM when completing

you're

transactions.

making

Follow

these

your transaction so people

general tips for your personal

waiting behind you won't have

safety.

an opportunity to see your PIN being entered or to view any

Stay

alert.

If

an

ATM

is

information. Similarly,

housed in an enclosed area,

fill out your deposit/withdrawal

shut the entry door completely

slips privately.

behind you. If you drive up to an ATM, keep your car doors

Don’t flash your cash. If you

locked and an eye on your

must count your money, do it

surroundings.

at the ATM, and place your

uneasy

or

If

sense

you

feel

something

cash into your wallet or purse

may be wrong while you're at

before stepping away. Avoid

an ATM, particularly at night or

making

www.pediain.com

excessively

large

11

ATM WITH AN EYE withdrawals. If you think you're

station

for

more

being followed as you leave

safety information.

personal

the ATM, go to a public area near

other

people

and,

if

FACIAL RECOGNITION

necessary, ask for help.

Save

receipt.

Your

developing such a model are

receipts provide a record of

keeping the time elapsed in

your transactions that you can later

reconcile

monthly

bank

with

your

statement.

If

you notice any discrepancies on

your

statement,

The main issues faced in

ATM



your bank as soon as possible. Leaving receipts at an ATM can also let others know how much money you've withdrawn and how much you have in your

the verification process to a negligible amount, allowing for an

appropriate

level

of

variation in a customer’s face when

compared

database

image,

to

the

and

that

credit cards which can be used at ATMs to withdraw funds are generally issued by institutions that do not have in-person

.

with the customer, and Guard your card. Don't lend your card or provide your PIN to others, or discuss your bank

with

friendly

strangers. If your card is lost or stolen,



your

bank

hence

no

opportunity

to

acquire a photo. Because the system would only attempt to match two (and later, a few) discrete images, searching through a large database of possible

immediately.

matching candidates would be unnecessary. Immediately

report

any

The

process

would effectively become an

crime to the police.

exercise in pattern matching,

the

which would not require a

Department

Of

Public

Security or your local police

www.pediain.com

great

deal

of

time.

With

12

ATM WITH AN EYE appropriate lighting and robust learning

software,

slight

In the case of credit card use

at

ATMs,

such

a

variations could be ed

verification system would not

for in most cases. Further, a

currently be feasible without

positive visual match would

creating an overhaul for the

cause the live image to be

entire

stored in the database so that

industry, but it is possible that

future transactions would have

positive

results

a broader base from which to

significant

fraud

compare if the original

achieved by this system might

image fails to provide a match

motivate such an overhaul.

–

thereby

decreasing

credit

card

issuing

(read: reduction)

false

negatives.

The last consideration is that consumers may be wary of the

When a match is made with

privacy concerns raised by

the PIN but not the images, the

maintaining

bank could limit transactions in

customers in a bank database,

a manner agreed upon by the

encrypted or otherwise, due to

customer when the

possible hacking attempts or

was opened, and could store

employee misuse. However,

the image of the for later

one could argue that having

examination by bank officials.

the image compromised by a

In regards to bank employees

third party would have far less

gaining

dire consequences than the

PINs

access

for

use

transactions, would

likewise

to in

customer fraudulent

this

system

reduce

that



images

information

of

itself.

Furthermore, since nearly all ATMs

videotape

customers

threat to exposure to the low

engaging in transactions, it is

limit imposed by the bank and

no broad leap to realize that

agreed to by the customer on

banks already build an archive

visually

of their customer images, even

transactions.

www.pediain.com

unverifiable

if they are not necessarily

13

ATM WITH AN EYE grouped

with



information.

could easily be compiled for the Windows XP environment and

the

networking

and

database tools will already be

SOFTWARE SPECIFICATION

in place. For most of the past ten years, the majority of ATMs used

worldwide

IBM’s

ran

now-defunct

Many financial institutions

under

are relying on Windows NT,

OS/2.

because of its stability and

However, IBM hasn’t issued a

maturity as a platform.The

major update to the operating

ATMs send database requests

system

years.

to bank servers which do the

banking

bulk of transaction processing

in

Movement

over in

six

the

world is now going in two

(linux.org.) This model would

directions: Windows and Linux.

also work well for the proposed

NCR, a leading world-wide ATM

system if the ATMs processors

manufacturer,

were not powerful enough to

recently

announced an agreement to

quickly

use Windows XP Embedded in

recognition algorithms.

its

next

generation

personalized

Windows

choose

from

SECURITY

In

the



improvement

thousands of components that

standards,

make

spearheading

up

Windows

Professional,

networking

XP

including

integrated

multimedia, and

management

facial

XP

Embedded allows OEMs to pick and

the

of ATMs

(crmdaily.com.)

perform

database

functionality.

of of

the security

MasterCard an

effort

is to

heighten the encryption used at ATMs. For the past few decades, many machines have used

the

Data

Encryption

This makes the use of off-the-

Standard developed by IBM in

shelf facial recognition code

the mid 1970s that uses a 56-

more

desirable

www.pediain.com

because

it

bit key. DES has been shown

14

ATM WITH AN EYE to be rather easily cracked,

verification – matching a live

however,

image to a predefined image

given

proper

computing hardware. In recent

to a claim of identity –

years, a “Triple DES” scheme

not in the process of facial

has been put forth that uses

evaluation – matching a live

three

image to any image in a

such

keys,

for

an

effective 168-bit key length.

database.

ATM manufacturers are now

environmental

developing newer models that

under which the verification

Triple DES natively;

takes place – the lighting, the

such

imaging system, the image

redesigns

may

make

Further,

the

conditions

them more amenable to also

profile, and the processing

including

environment – would all be

and

snapshot facial

cameras

recognition

controlled

within

certain

software, more so than they

narrow limits, making hugely

would

robust

be

in

regards

retrofitting

to

pre-existing

machines .

software

unnecessary facial

.One

recognition

leading algorithm

class is called image template based. This method attempts to capture global features of FACIALRECOGNITION

facial

images

TECHNIQUE:

templates.

into

What

facial

must

be

taken into , though, There

are

proposed

hundreds and

implementations

of

actual of

facial

recognition technology from all manner

of

vendors

for

are certain key factors that may

change

images:

across

live

illumination,

expression, and pose (profile.)

all

manner of uses. However, for the model proposed in this paper, we are interested only in

the

process

www.pediain.com

of

facial

The natural conclusion to draw, then, is to take a frontal

15

ATM WITH AN EYE image for the bank database, and to provide a prompt to the

The

first

and

most

, verbal or otherwise, to

important step of this project

face the camera directly when

will be to locate a powerful

the ATM verification process is

open-source facial recognition

to begin, so as to avoid the

program

need to for profile

feature analysis and that is

changes. With this and other

targeted at facial verification.

accommodations,

This

recognition

that

uses

program

local

should

rates for verification can rise

compilable

above

systems, including Linux and

90%.

A

system

can

on

be

multiple

examine just the eyes, or the

Windows variants, and should

eyes nose and mouth, or ears,

be customizable to the extent

nose, mouth and eyebrows,

of allowing for variations in

and so on

processing

power

of

the

.

machines onto which it would

The conclusion to be drawn

be deployed.

for this project, then, is that facial verification software is

We

will

then

need

to

currently up to the task of

familiarize ourselves with the

providing high match rates for

internal

use in ATM transactions. What

program so that we can learn

remains

its strengths and limitations.

is

to

find

an

workings

of

the

appropriate open-source local

Simple testing of this program

feature

will also need to occur so that

analysis

facial

verification program that can be

used

on

a

variety

we of

could

evaluate

its

effectiveness. Several sample

platforms, including embedded

images will be taken of several

processors, and to determine

individuals to be used as test

behavior

cases – one each for “”

protocols

for

match / non-match cases.

the

images, and several each for “live” images, each of which

OUR METHODOLOGY

www.pediain.com

16

ATM WITH AN EYE would

vary

pose,

lighting

conditions, and expressions.

functioning properly, they will be

tweaked

possible

as

much

to

as

increase

performance (decreasing the Once

a final

chosen,

we

simple

is

time spent matching) and to

develop

a

decrease memory footprint.

will

ATM

program.

program

This

black

box

program

will

Following that, the black

server as the theoretical ATM

boxes will be broken into two

with

components – a server and a

which

recognition

the

facial

software

will

client – to be used in a two-

interact. It will take in a name

machine network. The client

and , and then look in

code

a folder for an image that is

interface, ing all input

associated with that name. It

data to the server code, which

will then take in an image from

will handle the calls to the

a

separate

images

and

recognition generate

a

will

act

as



folder

of

“live”

facial

use

the

facial

further reducing the memory

program

to

footprint and processor load

level

required on the client end. In

match

recognition

a

software,

between the two. Finally it will

this sense, the thin client

use the match level to decide

architecture of many ATMs will

whether

be emulated.

or

not

to

allow

“access”, at which point it will terminate. All of this will be

We will then investigate the

necessary, of course, because

process of using the black box

we will not have access to an

program to control a USB

actual ATM or its software.

camera

attached

to

the

computer to avoid the use of Both pieces of software will

the folder of “live” images.

be compiled and run on a

Lastly, it may be possible to

Windows

add

XP

and

a

Linux

system. Once they are both

www.pediain.com

some

sort

of

DES

encryption to the client end to

17

ATM WITH AN EYE encrypt the input data and



decrypt the output data from

identification

the server – knowing that this

and scratch it on the

will

back

increase

the

processor

of

their

personal

number

their

cards

or

or

load, but better allowing us to

somewhere that a potential

gauge the time it takes to

thief can find, no more fear of

process.

having an cleaned out if the card is lost or stolen.

IRIS RECOGNITION: HOW THE SYSTEM WORKS. Inspite security

of

all

features,

technology

these a

new

has

developed.

Bank

been United

of

When a customer puts in a bankcard, a stereo camera locates the face, finds the eye

Texas became the first in the

and takes a digital image of

United

the iris at a distance of up to

States

recognition automatic providing

to

offer

iris

technology

at

teller the

machines,

customers

a

three

feet.

The

resulting

computerized "iris code" is compared

with

one

the

cardless, -free way to

customer will initially provide

get their money out of an ATM.

the bank. The ATM won't work

There's

if the two codes don't match.

no

card

to

show,

there's no fingers to ink, no

The entire process takes less

customer

than two seconds.

inconvenience

discomfort.

It's

or

just

a

photograph of a Bank United customer's eyes. Just step up to the camera while your eye is scanned. The iris

-- the

colored part of the eye the camera will be checking -- is unique to every person, more so than fingerprints. And, for the

customers

www.pediain.com

who

can't

The

system

works

equally well with customers wearing glasses or lenses and at night. No special lighting is needed. The camera also does not use any kind of beam. Instead, a special lens has been developed that will not only blow up the image of

18

ATM WITH AN EYE the

iris,

but

provide

more

When the system is fully

detail when it does. Iris scans

operational, a bank customer

are much more accurate than

will have an iris record made

other

for

high-tech

available

that

ID

systems

scan

voices,

faces and fingerprints.

comparison

when

an

is opened. The bank will

have

the

option

of

identifying either the left or Scientists

have

identified

250 features unique to each person's iris -- compared with about 40 for fingerprints -- and it remains constant through a person's life, unlike a voice or a face. Fingerprint and hand patterns

can

be

changed

through alteration or injury. The iris is the best part of the eye

to

use

as

a

identifier

because there are no known diseases of the iris and eye

right eye or both. It requires no intervention by the customer. They will simply get a letter telling them they no longer have to use the PIN number. And, scam artists beware, a picture of the card holder won't muster. The first thing the camera will check is whether the eye is pulsating. If we don't see blood flowing through your eye, you're either dead or it's a picture.

surgery is not performed on the iris. Iris identification is the most secure, robust and stable form of identification known to man. It is far safer, faster, more secure and accurate than DNA

testing.

Even

identical

twins do not have identical irises. The

iris remains the

same from 18 months after birth until five minutes after death.

www.pediain.com

firstdata.com Retail ATM Program Triton RL5000 Model The

Triton

automatic

RL5000 teller

model machine

(ATM) brings a superior level of performance, reliability and customization options to the embedded PC ATM industry.

19

ATM WITH AN EYE This

durable

ATM

solution

installation,

comprehensive

allows merchants to offer their

training, technical ,

customers the convenience of

service and reporting features.

an on-site ATM, which may

RL5000

lead

firstdata.com

to

increased

customer

traffic, increased sales volume

301-333

and

Retail ATM Program

an

additional

revenue

source. From small specialty

Triton RL5000 Model

stores to larger petroleum and

Features

convenience stores, the

JJMonitor

Triton RL5000 ATM is easily

- 10.4" VGA color screen

configured to meet the needs

JJAdvertising Potential

of merchants of all sizes.

- Decals and signage

The

Triton

RL5000

ATM

- Couponing

features the -friendly

- Display graphics

Microsoft® Windows® CE 5.0

JJPrinter

operating system and has a

- 80mm thermal printer with

large

graphics capabilities

capacity

for

storing

journal records or ing

JJKeypad

graphics

- Encrypting PIN Pad (EPP)

for

marketing

improved

efforts.

The

- PCI compliant

prominent

JJCommunications

10.4" LCD display and sleek

- Standard dial

profile make the ATM an

JJSmall Footprint

eye-catching

component

of

any merchant location. Many

other

on-site

JJBacklit Topper -

ATM

Mid-topper

standard

increases visibility

programs can be expensive

JJCard Reader

and

- Dip standard

difficult to implement, while

JJADA Compliant

the Triton RL5000 ATM offered

- Americans with Disabilities

by First Data is a complete

Act

ATM solution that includes

www.pediain.com

20

ATM WITH AN EYE (ADA)

compliant

and

audio

Relative humidity: 20% to 80%

compliant

non-condensing

JJTechnical Specs

Power consumption: 2.0A @

- RISC microprocessor

115 VAC at 60 Hz;

- Microsoft® WIndows® CE 5.0

1.0A @ 230 VAC at 50 Hz

-

A

128

MB

flash

memory,

Global

Leader

in

expandable

Electronic Commerce

- 128 MB RAM

First Data powers the global

- Multilingual capabilities

economy by making it easy,

- Cassette options available

fast and secure for people and

JJSecurity

businesses around the

- Electronic lock standard

world

Benefits

services using virtually any

JJSupplementary

revenue

source (potential profit from

ATM

buy

goods

and

form of payment. Serving millions of

surcharge

and/or

interchange) JJHigher

to

merchant locations and thousands of card issuers, we

sales

volume

from

have the expertise and

more in-store purchases

insight to help you accelerate

JJIncreased store traffic

your business. Put our

JJSuperior

intelligence to work for you.

customer

convenience

For more information,

JJImproved customer loyalty

your

Dimensions

First Data Sales Representative

Height: 64.9" (with mid-topper)

or visit firstdata.com.

Width: 16.9"

© 2009 First Data Corporation.

Depth: 25.6"

All

Weight: 245 lbs.

trademarks, service

Color: Bayou Bronze

marks

Operating Environment

referenced in this material are

Temperature

range:

40°C / 50° to 104°F

www.pediain.com

10°

to

rights

and

reserved.

trade

All

names

the property of their respective owners.

21

ATM WITH AN EYE CONCLUSION:

Bone,

Mike,

James We thus develop an ATM

L.,

Wayman, and

Duane.

Dr.

Blackburn, “Evaluating

model that is more reliable in

FacialRecognition Technology

providing

for Drug Control Applications.”

security

by

using

facial recognition software. By

OND

International

keeping the time elapsed in

Counterdrug

Technology

the verification process to a

Symposium: Facial Recognition

negligible amount we even try

Vendor Test. Department of

to maintain the efficiency of

Defense

Counterdrug

this ATM system to a greater

Technology

Development

degree. One could argue that

Program Office, June 2001.

having

the

image

compromised by a third party

Gross, Ralph, Shi, Jianbo, and

would

Cohn, Jeffrey F. “Quo vadis

have

far

consequences

less than

information

dire the itself.

Face

Recognition.”

ThirdWorkshop on Empirical

Furthermore, since nearly all

Evaluation

ATMs

Computer

videotape

customers

engaging in transactions, it is

Methods Vision.

in Kauai:

December 2001.

no broad leap to realize that banks already build an archive

Penev, Penio S., and Atick,

of their customer images, even

Joseph

if

Analysis:

they

are

grouped

not

necessarily

with



information.

J.

“Local A

StatisticalTheory

General for

Representation.” Computation

BIBILOGRAPHY:

Feature

Object Network:

in

Neural

Systems, Vol. 7, No. 3, pp. 477-500, 1996.

All, Anne. “Triple DES dare you.”

ATM

19 Apr. 2002.

www.pediain.com

Marketplace.com.

Wrolstad, Jay. “NCR To Deploy New Microsoft OS in ATMs.”

22

ATM WITH AN EYE CRMDailyDotCom.

29Nov.

2001

www.pediain.com

23