Biometrics

Introduction:

Thousands of years earlier physiological parameters such as scars, complexion,

eye color, height, moles, etc are used to identify the individuals. Later, in the nineteenth

century physical features and characteristics are used to identify criminals. This resulted

in a variety of measuring devices being produced. The idea of measuring individual

physical characteristics seemed to stick and the parallel development of fingerprinting

became the international methodology among police forces for identity verification.

With this background, it is hardly surprising that for many years a

fascination with the possibility of using electronics and the power of microprocessors to

automate identity verification had occupied the minds of individuals and organizations

both in the military and commercial sectors. After the September 11th terrorist attack,

Biometrics has gained prominence as being more reliable than current technologies, for

people identification.

Definition:

Biometrics involves identifying a person based on a unique physical

characteristic that is different from any other person. Biometrics can be either ‘innate’

such as fingerprints, face or iris; or ‘behavioral’ such as handwriting, gait or typing

Style. Biometric characteristics are measured using sensors that produce data values that

can then be processed by a computer using specialized algorithms for analysis and

comparison. Biometric system is the integrated biometric hardware and software used to

conduct biometric identification or verification.

Biometric devices and systems have their own operating methodology, there

are some generalisations for biometric systems implementation as shown in the fig.

[A] Enrollment: The process whereby a user’s initial biometric sample or samples are

collected, assessed, processed, and stored for ongoing use in a biometric system. These

samples are referred as templates. A poor quality template will often cause considerable

problems for the user, often resulting in a re-enrolment.

which may accommodate many thousands of individuals. The possible options are as

follows; Store the template within the biometric reader device or remotely in a central

repository or on a portable token such as a chip card. The last one is an attractive option

for two reasons. Firstly, it requires no local or central storage of templates and secondly,

the user carries their template with them and can use it at any authorized reader position.

[C] The network. Networks may rely on the vendor’s systems design and message

functionality, together with their own software. Or the user can himself design

networking, message passing and monitoring system ,taking advantage of the recent

generic biometric API’s and accessing the reader functions directly. This provides

absolute flexibility and control over systems design.

[D] Verification: The verification process requires the user to claim an identity by

either entering a PIN or presenting a token, and then verify this claim by providing a live

biometric to be compared against the claimed reference template. There will be a

resulting match or no match accordingly. A record of this transaction will then be

generated and stored, either locally within the device or remotely via a network and host

(or indeed both).

[E] Transaction storage: Some devices will store a limited number of transactions

internally, scrolling over as new transactions are received.

[F] The network (again): Here the network handles transactions, which is of critical

importance in some applications.

Popular Biometric Methodologies:

1.Finger Print Identification:

identification is the oldest method, which has been successfully used in numerous

applications. A fingerprint is made of a series of ridges and furrows on the surface

of the finger. The uniqueness of a fingerprint can be determined by the pattern of

ridges and furrows as well as the minutiae points. Minutiae points are the

locations on your fingerprint where the ridges will stop or split into two, or

intersect. Fingerprint matching techniques can be placed into two categories:

minutiae-based and correlation based. In minutiae-based matching a fingerprint is

initially enrolled into an archive. Instead of storing the entire image of the

fingerprint, only the minutiae points are kept. An algorithm is used to translate the

minutiae points into a code, which is called the template. Later, when the person

has their fingerprint scanned, the minutiae points are recognized, placed through

the algorithm and the code is compared to the template. But it is difficult to

extract the minutiae points accurately when the fingerprint is of low quality.

Local ridge structures cannot be completely characterized by minutiae. The

correlation-based techniques require the precise location of a registration point

and are affected by image translation and rotation. So an alternate representation

of fingerprints is tried, which will capture more local information and yield a

fixed length code for the fingerprint.

• Fingerprint Classification: An automatic recognition of people based on

fingerprints requires that the input fingerprint be matched with a large number of

fingerprints in a database . The input fingerprint is required to be matched only

computational complexity.

An algorithm classifies fingerprints into five classes namely, whorl, right loop, left loop,

arch, and tented arch, as shown in figure. The algorithm separates the number of ridges

present in four directions (0 degree, 45 degree, 90 degree, and 135 degree).

2.Speaker Recognition:

It involves recognizing people based on their voices. Systems plot

frequencies that the voice produces every 1/100th of a second to form a "waterfall." The

"waterfall" is a three-dimensional image that is a map of the voice for one second. A line

graph of curved and straight arrows note the transitions between phonemes, the smallest

units of speech. A database stores this information until it is needed. Voiceprint

determines resonance of the cavities in the throat with vocal-tract geometry from a voice

sample. Different languages, different voice samples, stuffy noses etc cannot fool this

method. There are two types of models that have been used extensively in speaker

verification and speech recognition systems: stochastic models and template models. The

stochastic model treats the speech production process as a parametric random process and
assumes that the parameters of the underlying stochastic process can be estimated in a

precise, well-defined manner. The template model attempts to model the speech

production process in a non-parametric manner by retaining a number of sequences of

feature vectors derived from multiple utterances of the same word by the same person. A

very popular stochastic model for modeling the speech production process is the Hidden

Markov Model (HMM). The model is a doubly embedded stochastic process where the

underlying stochastic process is not directly observable (it is hidden). The HMM can only

be viewed through another set of stochastic processes that produce the sequence of

observations. Thus, the HMM is a finite-state machine, where a probability density

function p(x | s_i) is associated with each state s_i. The states are connected by a

transition network, where the state transition probabilities are a_{ij} = p(s_i | s_j). A fully

connected three-state HMM is depicted in figure

For the Hidden Markov models, the matching score is the probability that the model

generated a given set of feature vectors.

3.Facial Recognition:

• Eigenfaces: Developed by Viisage Technology. First, a camera captures the

image of a person. The face is mapped into a variety of coefficients depending on

the features, which becomes 128 digits, otherwise called a "vector." The vector is
 the code for the person. When the person presents him/herself later to be

identified, he/she must present a PIN or bar code and then allow the camera to

take a picture. The difference between the face's vector and the database's vector

for the PIN or bar code are placed on a coordinate system called "face space" and

the differences between the two of them are determined.If distance between the

images is small enough, access is granted, else denied

Local Feature Analysis: Local Feature Analysis (LFA) is another algorithm. First, the

camera takes a picture of the face and cuts out the background and other faces. LFA is

based on idea that all faces are made of building blocks. However, every face uses

different building blocks and puts them together in a unique manner. LFA translates the

building blocks into a code called "face print." The "face print" is stored in a database for

identification. LFA is also used for verification.

Neural Networks: Neural Networks is a slight bit more complicated than the

"eigenfaces" method. A series of neurons, or processing units, are connected together. A

programmer sets the rules of how the neurons recognize patterns. If the actual output is

greatly different than the computed output, then the neurons will adjust itself to fit the

situation. More situations lead to more shifts in neurons, which lead to better results.

MultiBiometrics:
A biometric system, which relies only on a single biometric identifier in making a

personal identification, is often not able to meet the desired performance requirements.

We introduce a multimodal biometric system, which integrates face recognition,

fingerprint verification, and speaker verification in making a personal identification. This

system takes advantage of the capabilities of each individual biometric. A multimodal

biometric system integrates face recognition, fingerprint verification, and speaker

verification in making a personal identification as shown in figure.

Applications:

• Time Attendance

• Access control
 • Identification card and Immigration checks

• Police records, Patient management in hospitals

• Customer identification, Loyalty programs

• Security systems and Preventing identity theft

• Membership management in clubs, libraries etc.

Future Applications:

 ATM machine use: Most of the leading banks have been

experimenting with biometrics for ATM machine use and as a general

means of combating card fraud.

 Workstation and network access: Many are viewing this as the

application, which will provide critical mass for the biometric industry and

create the transition between sci-fi device to regular systems component,

thus raising public awareness and lowering resistance to the use of

biometrics in general.

 Travel and tourism: There are multi application cards for travelers

which, incorporating a biometric, would enable them to participate in

various frequent flyer and border control systems as well as paying for

their air ticket, hotel room, hire care etc.

 Telephone transactions: Many telesales and call center managers have

pondered the use of biometrics.

Real-time Application:

U.K Airport Tests Passenger Eye ID’s: Heathrow is the first UK airport to carry out a

large-scale trial of the iris recognition technology, which examines a passenger's eye,
rather than their passport as they go through immigration control. The aim is to speed up

the movement of passengers through the terminal and detect illegal immigrants. Each

passenger will have an image of one of his or her eye's iris stored on computer. Instead of

showing their passport on arrival they will go into a kiosk where in seconds a camera will

check that the pattern of their iris matches computer records. It is hoped the technology

could have future security benefits, with UK airports still on alert following 11

September. The entire procedure only takes a few seconds and there is no contact with

the body or with lasers or other potentially harmful light sources. Passengers taking part

are being asked to carry their passports during the trial period should immigration

officials want to check their details.

Benefits:

 Increased security when controlling access to confidential data and IT systems

 Reduced risk of fraudulent use of identity by employees

 Enhanced user convenience

 Increased costs savings due to reduced password maintenance costs

 User convenience due to no passwords to remember or reset and faster login

 Privacy – ability to transact anonymously

Demerits:

An automatic personal identification system based solely on fingerprints or faces

is often not able to meet the system performance requirements.

In case of face recognition, face will sometimes change with time or injury, and

that poses a problem

Fingerprint verification is reliable but inefficient in database retrieval.

Some voice recognition systems has some problems since the voice changes with

a human’s mood and illness and background noise poses some problems.

Biometrics In Future:

In future with Biometrics, one can go to shopping without money purse or

credit card. It just needs to place his/her finger on the reader device and can debit money

from their bank accounts. In credit cards system if a mother and daughter have a joint

account, then they cannot know how much each one uses. But with biometrics mother

will know her daughter secretly buying beer bottles. The biometrics market is expected to

grow to $1.8 billion by 2004.

Conclusion:

Biometrics makes automated use of physiological or behavioral characteristics to

determine or verify identity. Finger biometrics are most popular and one of the most

accurate and cost effective solutions. Hand geometry, signature-scan, keystroke-scan,

palm-scan are some more biometric technologies in use. With Biometrics there is no

problem of ID being stolen. Many institutions and organizations are trying to use this

technology.

ABSTRACT

The automated use of physiological or behavioral characteristics to

determine or verify identity. To elaborate on this definition, physiological biometrics are

based on measurements and data derived from direct measurement of a part of the human

body. Fingerprint, iris-scan, retina-scan, hand geometry, and facial recognition are
leading physiological biometrics. Biometric system is the integrated biometric hardware

and software used to conduct biometric identification or verification. A generalized

biometric system is expressed nearly with figure.

Various biometric technologies are fingerprint identification, speaker

recognition, facial recognition, hand geometry, signature-scan, keystroke-scan, palm-

scan, etc. Among these popular recognition techniques like fingerprint recognition , facial

recognition and speech recognition are clearly explained. An automatic personal

identification system based solely on one methodology often cannot meet the system

performance requirements. So a combination of two or more methodologies is used to

achieve required performance, which is called multibiometrics. In this paper we

explained an example of multibiometric system, which combines fingerprint

identification, speech recognition and facial recognition. A neatly sketched figure is used

to describe the process in brief.

Biometrics has wide area applications; Most of them are covered in this paper.

Also some of the future applications like ATM machine, workstation and network access,

travel and tourism and telephone transactions are given.

Heathrow is the first UK airport to carry out a large-scale trial of the iris

recognition technology, which examines a passenger's eye, rather than their passport as

they go through immigration control. This real-time application is explained. There are
many benefits with biometrics. And so the merits and demerits of it are listed. At last we

predicted how biometrics changes lifestyle in future.

The structure of this document together with all pictures is original. The text

expresses our own understanding on the concepts and is based on the documents referred

from selected books.

Keywords: Biometrics, biometric system, enrollment, template storage, network,

fingerprint identification, fingerprint matching, fingerprint classification, Speaker

recognition, waterfall, stochastic model, template model,, Hidden Markov Model, Facial

recognition, Eigenfaces, Local feature analysis, neural networks, multibiometrics.