You are on page 1of 25

1

CHAPTER 1

INTRODUCTION

1.1 GENERAL

The human authorization plays an important role in the security


system. Every human consists of biometric features. The features that are
extracted in the form of a matrix are called biometrics. The biometric system
uses any of the human resources to provide the authentication. It gives human
identity based on physiological and behavioural characteristics.

The biometric system consists of sensor, signal processing


algorithm, data storage, data matching algorithm and decision making stages.
The sensors are used to capture the biometric resources of the human. The
signal processing algorithm extracts the unique features and creates the
template from the captured biometric signal. The templates are stored in the
data storage. The new templates are compared with the data base templates
using matching algorithm. The decision making process is carried out using
the matching result.

The implementation of a biometric system should consider the


installing location, safety of the device, task to be performed, number of users
and user privileges. The system has to assure the properties such as high
security, uniqueness, universality, user accessibility and robustness.
2

1.2 NEED FOR BIOMETRIC SYSTEM

A highly securable biometric based authorization is required due to


the increase in security violations, criminal activities and transaction scams.
All biometric resources will not be suitable for all places. In such a case,
instead of a single biometric resource more than one resource is required.

1.3 APPLICATIONS OF A BIOMETRIC SYSTEM

The biometric system is used for the direct human authorization


and also in the electronic security system. The biometric system is used in
government and private organizations, military and commercial applications.
The biometric systems are used in National Unique Identification card
(NUID), Automated Teller Machine (ATM), employee authorization,
employee in and out time attendance system, criminal identification, law
enforcement, human authorization in restricted places, banking applications,
passport, driving license, Air/rail transport reservation system, health and
social services.

The electronic security based biometric applications assures the


secret financial transactions and maintains the personal data information. The
electronic transaction improves the growth of the country. It is used in
electronic banking which provides the confidential money transaction. It is
also used in the computer network system to manage the workstation, domain
access, logging on, logon privileges, remote access and web security.

1.4 BIOMETRIC IMPLEMENTATION IN INDIA

Our Indian government is in the process of implementing the NUID


under the project name called Aadhaar. In this project the three biometric
sensors are used to enroll the biological information of the people. An iris
3

scanner device is used to collect the left and right eye information. A web
camera is used to collect the facial information. The finger print scanner
device is used to collect the fingers information. The population of India is
402,390,930 in the recent census. Almost 34% of enrollment process is
completed. The state wise enrollment statistics details are shown in the
Figure 1.1.

Figure 1.1 Biometric enrollment process in India

In banking sector almost all the nationalized banks use the


biometric system in the ATM machines. The banks also collect the NUID
number from their customers. The private industries and organizations use the
biometric system for the purpose of employee enrolment and time
management. The project called Electronic Shakti (e-Shakti) implemented by
4

state government of Bihar works on the biometric system. In this project the
employee attendance, entry and exit time are enrolled for wages. Andhra
Pradesh state government has implemented the biometric system in the ration
card.

The educational institutions use the biometric system for students


and staffs presence for time management. The attendance and access control
system are developed using the finger print based biometric system. In health
care application, the biometric systems are implemented in infant protection
system, biometric patient identity management, new born baby foot print
identification, hospital tracking and trace, building security with smart
protection using Radio Frequency Identification (RFID) technology.

The commercial applications such as food coupons, ticket counters,


membership verification, self service areas and retail shops use biometric
based identity. It is used in the law and enforcement system for criminal
tracking and identification, border control security solutions, civil
identification and common integrated police applications.

1.5 TYPES OF BIOMETRIC SYSTEM

The biometric systems are classified into two groups. First one is
based on the physiological parameters and the other one is based on
behavioural characteristics of human. The physiological biometrics work on
the physical parameters of the human. The behavioural biometrics work based
on the movement or action of the human (Bhattacharyya et al 2009).

The various biometric resources are finger print, hand geometry,


palm print, ear shape, nose shape, iris, retina, Sclera, face, Deoxyiribo Nucleic
Acid (DNA), thermo grams, skin reflection, lip motion, body odour, vascular
5

pattern, vein pattern, sweat pores, gait, key stroke, signature, voice, foot print
and foot dynamics [13].

1.5.1 Finger Print Biometric System

The finger print scanner is used to collect the finger impression.


The finger impression will vary from person to person. A sample of finger
print scanner and the finger impression are shown in the Figure 1.2. The
finger print device is mostly used for attendance in time management systems.

Figure 1.2 Finger print scanner and impression

Advantages:

It is very simple to use.

Lesser time is required to enroll the finger impression.

Drawbacks:

Duplicate finger impressions may be used by the criminals.

Causes health hazard by spread of diseases from one person to


another person.

Due to ageing, the repetition may be affected.


6

1.5.2 Hand Geometry

The hand geometry scanner scans all the finger impressions. The
hand geometry biometric device is shown in the Figure 1.3. It is the advanced
version of finger print scanner. The instrument recognizes the person based on
all the finger impression. It has similar advantages and drawbacks as that of
finger print biometrics.

Figure 1.3 Hand geometry scanner and fingers impressions

1.5.3 Palm Print Biometrics

The palm print biometric device scans the palm of the human hand.
The palm scanner and the scanned palm impression are shown in the Figure
1.4. It is used in the multimodal biometrics system.

Figure 1.4 Palm biometric device and impression


7

Advantages:

Scanning and enrollment process is very simple.

It varies from human to human.

Drawback:

Due to ageing, the palm impression may alter.

Duplication is possible.

1.5.4 Facial Biometrics

The face biometric device captures the facial image using the
camera device as shown in the Figure 1.5. It analyzes the shapes of facial
parts such as eye position, nose position, lip and ear position. It does not have
direct contact with the human face.

Figure 1.5 Face biometric device

Advantages:

The device does not have any physical contact with the
human.
8

The process and enrollment is easy.

The validation time is very less.

Drawbacks:

It is not suitable in noisy environment such as more than a


person stand before the device and poor lighting.

The capturing of image is affected by glasses, hats, hair styles.

Due to facial expressions such as emotion and pose, the


performance is affected.

The input expression has to be maintained during validation


process.

1.5.5 Ear and Nose Biometrics

The captured ear and nose images are shown in the Figure 1.6. The
ear and nose shapes are taken as the feature. It does not have direct contact
with the human.

Figure 1.6 Shape of nose and ear biometrics

Advantages:

Images are easily captured.


9

Handling of the device is easy.

Used as a part of face biometrics.

Drawbacks:

Separating nose and ear makes a difficult task from the whole
image.

Facial reaction may affect the performance of the device.

1.5.6 Vein Biometrics

The vein biometrics is the advanced version of palm biometrics.


The Infra Red (IR) rays are passed through the palm to capture the vein
pattern. The vein pattern scanner device and vein patterns are shown in the
Figure 1.7.

Figure 1.7 Vein biometrics scanner

Advantages:

Enrollment process is very simple.

The pattern provides unique property.


10

Drawbacks:

The scanner machine has direct skin contact and brings health
related issues.

Due to ageing, the vein pattern may alter.

1.5.7 Lip Biometrics

The captured lip images are shown in the Figure 1.8. The shape of
lip contour is measured and taken as the feature. It is one among the features
of face biometrics. The lip colour and shape are some of the features of lip
biometrics.

Figure 1.8 Lip biometric source

Advantage:

Capturing process is simple.

Drawbacks:

Facial reaction may change shape.

The face hair may affect the performance.


11

1.5.8 Thermo Gram Biometrics

The heat pattern of the human body is captured using the IR


camera. The performance of this pattern is unique for every human. The heat
pattern of a human face is shown in the Figure 1.9.

Figure 1.9 Thermo gram image

Advantages:

It provides unique pattern for every human.

It is sensitive and accurate.

Drawbacks:

The sensor and devices are more expensive.

The uniqueness depends on the heat pattern.

It may lead to skin related issues.

1.5.9 Body Odour

The body odour gets varied due to organically variable chemical


structures of a human. The device consists of chemical vapour sensor array
12

which detects the odour feature. The body odour device is shown in the
Figure 1.10.

Figure 1.10 Body odour sensor device

Advantage:

Capturing process is easy.

Drawbacks:

The device is costly.

Capturing of odour depends on the atmospheric condition.

1.5.10 Skin Reflection Biometrics

The IR rays are passed inside the human skin. The light rays
penetrate the human body and IR receiver receives the scattered rays as
shown in the Figure 1.11. The received patterns are analyzed for the
biometrics.
13

Figure 1.11 Skin biometrics

Advantages:

The obtained pattern will have unique value.

The pattern capturing is easy.

Drawbacks:

Absorption spectrum value depends on the individual.

The artificial skin colour may affect the performance of the


system.

1.5.11 Gait Sensor

The gait sensor is used to sense the movement of human beings.


The gait biometric system is shown in the Figure 1.12. The gait device is
fixed on the human left leg and measurements are carried out.
14

Figure 1.12 Gait biometrics

Advantages:

It is a soft biometric system.

The human physic will not be affected.

Drawbacks:

More time is required to measure the gait information.

More number of samples is required to get the accurate


results.

1.5.12 Signature Biometrics

The signature is scanned using the image scanner device. The


feature points are fixed on the signature image. The example of signature
image is shown in the Figure 1.13.

Figure 1.13 Signature biometrics


15

Advantages:

Enrollment process is easy.

Easy to verify the system.

Drawback:

There are possibilities for duplication.

1.5.13 Keystroke Biometrics

The keystroke biometric is a combination of keyboard and


software. It measures the key switch press and release time as the feature of
biometrics. The view of key stroke biometrics is shown in the Figure 1.14.

Figure 1.14 Keystroke biometrics

Advantages:

It is a very easy process.

Cost of the system is very less.

Drawbacks:

Repetition is required for identification.

More numbers of features are required for a single person.


16

1.5.14 Audio / Voice Biometrics

In the voice biometrics, voice pattern is captured and feature points


are fixed on the voice pattern. The sample of voice pattern obtained for the
recognition is shown in the Figure 1.15.

Figure 1.15 Voice pattern

Advantages:

The audio can be easily captured.

The system is easily adaptable and cost of the system is very


less.

Drawbacks:

Throat infection may affect the voice.

Possibilities for duplication are more.

1.5.15 DNA

The blood samples are collected from the human and the chemical
test is carried out. The human body cell is physically analyzed. Based on the
analysis the decision is made.
17

Advantage:

Identification accuracy is more even in the non living


condition.

Drawbacks:

Blood sample is required for the processing.

It is more expensive one.

1.5.16 Foot Print and Foot Dynamics

The foot print impression is taken and the feature is extracted from
the foot print. The foot print scanner is used to scan the foot impression.

Advantages:

Capturing process is easy.

The new born baby is identified using foot print.

Drawback:

Due to ageing, the foot dimensions get varied.

1.5.17 Vascular pattern

The vascular patterns are captured from the bottom side of the leg.
The IR rays are passed to sense the vascular patterns.

Advantage:

The pattern capturing is easy.


18

Drawbacks:

The performance will be affected by the hair.

Due to vascular diseases, the pattern may get changed.

1.5.18 Eye Biometrics

The human eye consists of retina and iris biometric resources. The
human eye is one of the sense organs used to provide the vision. The
complete optical system of the human is shown in the Figure 1.16. The outer
most layer of the eye is composed of cornea and sclera. The cornea is smaller
in size and located in the front part of eye. The sclera is the larger part and
covers the remaining part of the eye (Guang-zhu et al 2007). The cornea and
the sclera are connected by the ring called limbus. The middle layer consists
of the choroid, ciliary muscle and iris.

Figure 1.16 Human eye structure

The inner most layer consists of retina. It is not directly visible and
viewed using ophthalmoscope equipment. The retina consists of optic disk,
macula, fovea and blood vessels. The retina vessels carry the blood
19

circulation inside the eye and receive the blood from the choroid. The lens is
located between the iris and the retina. The lens reflects the received light.
The function of the lens is similar to the camera lens.

The aqueous humour is filled in the anterior chamber and the


posterior chamber. The anterior chamber is located between the cornea and
iris. The posterior chamber is located between the iris and lens. The ciliary
muscle is triangular in shape. The lens is flexible and it balances on the ciliary
muscle. The ciliary muscle covers the vitreous body. The front portion of the
eye consists of iris. There are two types of eye biometrics as follows,

Retina Biometrics

Iris Biometrics

1.5.18.1 Retina biometrics

The retina of the eye is shown in the Figure 1.17. The retinal scan is
performed by casting an unperceived beam. The low-energy IR light is passed
into a persons eye. This beam of light traces a standardized path on the
retina. The retinal blood vessels are more absorbent of this light energy. The
amount of light reflection varies during the scan.

Figure 1.17 Retina


20

The pattern variations are converted to computer code and stored in


a database. Retinal scan captures the pattern of the eyes blood vessels. Retina
as a biometric has certain advantages when compared to other biometrics. It is
very secure and uses a stable physiological feature.

Advantages:

It provides unique feature in every human eye.

The left and right retinas are different in nature.

Twins have non identical retina.

The retina will not be affected by environmental changes.

Drawbacks:

It is a very difficult process to capture the retina image.

The retina structure gets changed due to the diabetic


retinopathy.

1.5.18.2 Iris biometrics

A digital camera is used to capture the iris. The digital camera is


placed at a certain distance and the image of the iris is captured. In this system
more amount of eye information is collected along with the iris. The iris
image is shown in the Figure 1.18.

Figure 1.18 Iris image


21

Figure 1.19 Iris capturing with digital camera

Figure 1.20 Single iris scanner device

Figure 1.21 Dual iris scanner device

The digital camera system with iris capturing system is shown in


the Figure 1.19. In this system the distance between the camera and human
face should be properly maintained. The single iris image scanner device
22

shown in the Figure 1.20 is used to scan either the left or right eye. The dual
iris scanner device is shown in the Figure 1.21, which scans the left and right
eye simultaneously. If the IR light race passed on human eye over 10 seconds
with a minimum distance of 760 nm may cause the damage of iris cell and
swelling on eye (Nikolas & Margaret 2011). Normally capturing distance is
should be above this range.

1.6 FEATURES OF IRIS BIOMETRICS

The formation of iris begins during the third month of


gestation. The complete unique iris pattern is formed during
the 8th month of gestation. After few years of life, the colour
pigmentation takes place in the iris. The biometric enrollment
is carried out after 3 years of a babys birth.

The eye organ is highly protected. Hence, throughout the life


there will be no change in iris structure (Daugman 2003).

The iris patterns are complex and random in nature which


provides the uniqueness among the irises.

The eye is externally visible and easily captured from a


distance of certain meters.

Iris patterns are not genetically determined (Daugman 2004).

The Left and right eyes are unique.

Twins have non identical iris.

If the eye is transferred to the other human, the iris structure


gets changed and it will identify the source person from whom
the eye is transferred. Hence, the iris biometrics provides a
perfect information security (Daugman & Downing 2013).
23

1.7 CHALLENGES OF IRIS BIOMETRICS

The few portions of eye lids and eye lashes are located at the
top and bottom of iris, which are the hazards for iris
processing (Bowyer et al 2008).

The flash light impression may be present in the iris.

Blinking of eye may affect the iris capture performance


(Baker et al 2010).

Physical cross verification is not possible.

1.8 MOTIVATION

Iris biometric system provides maximum advantages when


compared with the other biometrics for providing human identity. A highly
protected human system is achieved by iris biometrics.

The current ongoing project of Indian government named


aadhaar collects the peoples left and right eye information for generating
Unique Identification (UID). The collected information may be used for
some other purpose in future and it may support by the proposed work.

The effective iris boundary detection, segmentation, feature


extraction and analysis algorithm is required to extract iris region in the iris
hazard environment and unique identification.

1.9 OBJECTIVE OF THE THESIS

To develop the iris authorization algorithm to achieve higher degree


of accuracy and reduced error rates. In this study of research work, the
objectives are to be obtained using the following.
24

To adopt the suitable localization method to get hazard free


iris segmentation.

To select and fit the appropriate enhancement techniques to


project the hidden pixels.

To implement the suitable feature extraction, feature


dimension reduction technique and neural network technique
to extract the unique feature and authorization.

1.10 ORGANIZATION OF THE THESIS

The thesis is organized in seven chapters. Introduction to biometric


system and its needs, motivation and objectives of the thesis are described in
chapter-1.

Chapter-2 discusses on the literature part of the research work. The


literature covers the various parts of iris biometrics such as general iris
biometrics, iris data base, boundary detection, segmentation, enhancement,
feature extraction, dimensionality reduction, neural network and hardware
implementations.

Chapter-3 deals with the iris boundary detection. The discussion is


made on filter selection process, morphological operations, binary operators,
pupil boundary detection and iris boundary detection process.

Chapter-4 discusses iris segmentation and enhancement. This


chapter deals with the extraction of iris region from the eye. The elimination
of eye lashes and eye lids on the iris are discussed. The enhancement
technique adopted for the iris recognition process is also discussed.
25

Chapter-5 deals with the feature extraction and dimensionality


reduction. The feature extraction from the enhanced iris boundary is
discussed. The adaptation of 2D complex dual tree wavelet transform feature
extraction for iris is discussed. The adaptation of Multilinear Principle
Component Analysis (MPCA) technique for dimensionality reduction is
discussed.

Chapter-6 deals with the neural network for implementing iris


feature. The various biometric standards are evaluated to check the
performance of the proposed work. The discussion is also made on FPGA
based hardware implementation.

Conclusion and future scope of the proposed research work are


explained in chapter-7.

1.11 SUMMARY

The biometric based human security was discussed. The


classification of biometrics and types of biometrics were explained. The
various applications of biometrics were shortlisted. The ongoing biometric
project works in India were discussed. The advantages and drawbacks of
various biometrics systems were discussed. The features and challenges of iris
biometrics were discussed. The motivation of research work and objective of
the thesis were described. The organizations of thesis chapter were
mentioned.

You might also like