System Requirement Specification

BioMetric Criminals Identification System
Submitted By:
Abdul Bari Malik 905-FBAS/BSSE/F09
Asif Sharif 909-FBAS/BSSE/F09

Project Supervisor:
Mr. Syed Muhammad Saqlain
Assistant Professor

Department of Computer Science and Software Engineering
Faculty of Basic and Applied Sciences
International Islamic University, Islamabad
2013

I n the name of Allah (SWT) who is most Beneficent and most
Merciful

Faculty of Basic and Applied Sciences
Department of Computer Sciences and Software Engineering

FINAL APPROVAL
Dated:
It is certified that we have read the project submitted by Abdul Bari and Asif Sharif and it
is our judgment that this project is of sufficient standard to warrant its acceptance by
International Islamic University, Islamabad for the Bachelors Degree in Software
Engineering.
COMMITTEE
External Examiner

Mr.Muhammad Nadeem
Assistant Professor,
DCS & SE, FBAS,

Internal Examiner
Mr. Zulqarnain Hashmi
Lecturer,
DCS & SE, FBAS,

Supervisor
Mr. Syed Muhammad Saqlain
Assistant Professor
DCS & SE, FBAS,

BCIS Dissertation
i

A dissertation submitted to
DEPARTMENT OF COMPUTER SCIENCES
AND
SOFTWARE ENGNEERING
INTERNATIONAL ISLAMIC UNIVERSITY ISLAMABAD,
As partial fulfillment of the requirement of Bachelors degree in
Computer Sciences
BCIS Dedication
ii

DEDICATION
This thesis is dedicated to our parents who have been a great source of motivation and
inspiration and supported us all the way since the beginning of our studies.
BCIS Declaration
iii

DECLARATION
We hereby, declare that this project report, neither as a hole nor as a part there of has been
copied out from any source but in fact if any material has been used ,it is used as the
concept of original material taken from the web and all other references are mentioned for
subsequent sections. It is further declared that we have completed our work on the basis
of our personal efforts and under the sincere guidance of my teachers. If any part of this
part of work is proved to be copied out from any source, we will stand by the
consequences. No portion of Work presented in this report has been submitted in support
of any application for any degree or qualification of this or any other university.

Abdul Bari Malik
905-FBAS/ BSSE /F09
Asif Sharif
909-FBAS/ BSSE/ F09

BCIS Acknowledgment
iv

ACKNOWLEDGEMENTS
To Him belongs the dominion of the Heaven and the Earth, it is He who gives life and
death and He has power over all things (Al-Quran).
First of all we would like to thank the Almighty Allah, who is invincible and He, who
blessed us with the abilities to complete this project. The completion of this project was
only dependent on His blessings. Whenever we faced any problem we always looked
forward towards Allah The Almighty for His blessings to help us in standing large and be
determinant in the problem.
There are a number of people without whom this project might not have been written, and
to whom we are greatly indebted.
We Especially Thanks to our parents who helped us in our most difficult times and it is
due to their untiring effort that we are at this position.
We would also like to thanks all our Professors throughout our Bachelors course for
giving us this opportunity to learn and grow towards attainment of this degree.
Whenever one has a difficult task ahead, the first problem is how you climb and see
through the first stair because after the first step ahead, one always tries to move up and
up. So when we decided to take Biometric Person Identification using Face
Recognition as our final project this was our move towards the first stair and we were
fully supported to step up on the first stair by our teacher Mr. Syed Muhammad Saqlain
who later wore the gown of our project supervisor. Without his instructions and support
we wouldnt be able to move towards our destination.
Apart from the above honorable personalities we owe loving thanks to all our batch mates
who supported us and always gave us new ideas to include and were there as our
CRITICS, which helped us to climb till the last stair.
Finally, we would like to thank everybody who was important to the successful
realization of project, as well as expressing our apology that we could not mention
personally one by one.

Abdul Bari Malik
905-FBAS/ BSSE /F09

Asif Sharif
909-FBAS/ BSSE/ F09
BCIS Abstract
vi

PROJECT IN BRIEF

Project Title: BioMetric Criminals Identification System
Objective: To develop a person identification system along with
their profiles on the basis of their faces
Undertaken By: Abdul Bari
Asif Sharif

Supervised By: Mr. Syed Muhammad Saqlain
Date Started:
April 8,2013
Date Completed:
December 29, 2013
Language: JAVA
Tool(s) Used: Eclipse IDE
Microsoft Office 2010
Sparx Enterprise Architect 7.5
Visual Paradigm for UML 11

Operation System: Microsoft Windows7
System Used: Processor Dual Core, 2.80GH

BCIS Abstract
vii

ABSTRACT

This system is able to verify a persons identity through his/her facial features. This is
achieved by using Eigen faces, Principal Component Analysis (PCA) and OpenCV
Object Detection.

Persons are enrolled into the system by providing their personal information and frontal
face samples. These face samples are verified by detecting face in the face sample and
then preprocessing is performed on them. Once the persons are enrolled, the system is
trained on the preprocessed face samples using PCA (Principle Component Analysis).
Now it can be used to recognize persons, the person to be recognized comes in front of
camera, his/her face sample is taken, preprocessed and compared with the enrolled
images and the result is returned as matched or not matched.
BCIS Table- of-Contents
vii

Table of Contents
CHAPTER 1 ........................................................................................................................ 1
1. Introduction .................................................................................................................. 2
1.1. Problem Statement ................................................................................................ 2
1.2. Purpose .................................................................................................................. 3
1.3. Scope ..................................................................................................................... 3
1.4. Definition, Acronyms, and Abbreviations ............................................................ 3
1.5. Objective of the Project ......................................................................................... 3
1.6. Positioning ............................................................................................................. 3
1.6.1. Problem Statement ......................................................................................... 3
1.7. Main Project Achievements .................................................................................. 4
1.8. Product Overview .................................................................................................. 4
1.8.1. Product Perspective ........................................................................................ 4
1.8.2. Assumptions and Dependencies .................................................................... 4
1.8.3. Cost and Pricing ............................................................................................. 4
1.8.4. System Model ................................................................................................ 5
...................................................................................................................................... 5
1.9. Product Features .................................................................................................... 5
1.9.2. User Characteristics ....................................................................................... 6
1.9.3. Functional Requirement ................................................................................. 6
1.9.4. Organization of Report .................................................................................. 6
1.9.5. Tools and Technologies ................................................................................. 7
CHAPTER 2 ........................................................................................................................ 8
2. Literature Survey ......................................................................................................... 9
2.1. Projects Developed ................................................................................................ 9
2.1.1. ImageWare Software, Inc. ............................................................................. 9
2.1.2. VeriLook SDK (Face identification for stand-alone or Web applications) . 10
2.1.3. TrueFace ...................................................................................................... 10
2.1.4. FaceSDK ...................................................................................................... 10
2.1.5. Intelligent vision systems, Inc...................................................................... 11
2.1.6. FaceIt Pc....................................................................................................... 11
2.2. Techniques for Face Recognition ........................................................................ 12
viii

2.2.1. Geometric Features ...................................................................................... 12
2.2.2. Eigen Faces .................................................................................................. 12
2.2.3. Graph Matching ........................................................................................... 13
2.2.4. Fisher Faces (Linear Discriminate Analysis algorithm) .............................. 13
2.2.5. Template Matching ...................................................................................... 14
2.2.6. Neural Network Approach ........................................................................... 14
2.2.7. Stochastic Modeling..................................................................................... 16
2.2.8. N-tuple classifiers ........................................................................................ 16
2.3. Basic Concepts .................................................................................................... 18
2.3.1. Preprocessing ............................................................................................... 18
2.3.2. Object Detection .......................................................................................... 18
2.3.3. Face Recognition ......................................................................................... 19
3. Analysis...................................................................................................................... 29
3.1. Use Cases Model ................................................................................................. 29
3.1.1. Use Case Diagram........................................................................................ 29
3.1.2. Use Cases Descriptions in Brief .................................................................. 30
3.1.3. Use Case Description in Detail .................................................................... 31
3.1.4. System Sequence Diagrams ......................................................................... 35
3.2. Domain / Conceptual Model ............................................................................... 39
3.3. Activity Diagram ................................................................................................. 40
3.3.1. Admin Side Activity Diagram ..................................................................... 40
3.3.2. User Side Activity Diagram ......................................................................... 41
CHAPTER 4 ...................................................................................................................... 42
4. System Design ........................................................................................................... 43
4.1. Class Diagram ..................................................................................................... 44
4.2. Entity Relationship Diagram ............................................................................... 45
4.3. Sequence Diagrams ............................................................................................. 46
4.3.1. SD1: Enrollment .......................................................................................... 47
4.3.2. SD2: Preprocessing ...................................................................................... 47
4.3.3. SD3: Training............................................................................................... 48
4.3.4. SD4: Testing ................................................................................................ 48
CHAPTER 5 ...................................................................................................................... 49
5. Implementation .......................................................................................................... 50
ix

5.1. Tools and Technology ......................................................................................... 51
5.2. Deployment Diagram .......................................................................................... 51
CHAPTER 6 ...................................................................................................................... 52
6. Software Testing ........................................................................................................ 53
6.1. Why is Software Tested?..................................................................................... 55
6.1.1. To Improve Quality: .................................................................................... 55
6.1.2. For Verification and Validation (V&V) ...................................................... 55
6.1.3. For Reliability Estimation ............................................................................ 55
6.2. Software Testing Techniques .............................................................................. 56
6.2.1. Test Case Design.......................................................................................... 56
6.2.2. White Box Testing ....................................................................................... 56
6.2.3. Basic Path Testing........................................................................................ 56
6.2.4. Regression Testing ....................................................................................... 56
6.2.5. Control Structure Testing ............................................................................. 57
6.2.6. Black Box Testing........................................................................................ 57
6.3. Software Testing Strategies ................................................................................. 57
6.3.1. Unit Testing ................................................................................................. 57
6.3.2. Integration Testing ....................................................................................... 57
6.3.3. Function Testing .......................................................................................... 58
6.3.4. System Testing ............................................................................................. 58
6.3.5. Acceptance Testing ...................................................................................... 58
6.4. BPIS Testing ....................................................................................................... 58
6.5. Test case .............................................................................................................. 58
6.6. Deriving Test Cases ............................................................................................ 59
6.7. Test Case Description.......................................................................................... 60
6.7.1. Loading of Menu.......................................................................................... 60
6.7.2. Enrollment Test Case ................................................................................... 60
6.7.3. User Input handling...................................................................................... 61
6.7.4. Face Detection ............................................................................................. 61
6.7.5. Database Inserting ........................................................................................ 62
6.7.6. Training ........................................................................................................ 62
6.7.7. Testing.......................................................................................................... 63
7. Conclusion ................................................................................................................. 65
x

8. User Manual ............................................................................................................... 73
Pc Minimum System Requirements ............................................................................... 73
8.1. Main Window ...................................................................................................... 73
8.2. Enrollment Wizard Window 1 ............................................................................ 74
8.3. Enrollment Wizard Window 2 ............................................................................ 75
8.4. Recognition Window........................................................................................... 76
References .......................................................................................................................... 78

xi

Table of Figures
Figure 2.1: Flowchart of Face Recognition using Eigen Faces ......................................... 22
Figure 3.1: Use Case Diagram ........................................................................................... 29
Figure 3.2: SSD1 Enrollment ............................................................................................. 37
Figure 3.3: SSD2 Preprocessing ........................................................................................ 37
Figure 3.4: SSD3 Training ................................................................................................. 38
Figure 3.5: SSD4 Testing ................................................................................................... 38
Figure 3.6: Domain Model ................................................................................................. 39
Figure 3.7: Admin Side Activity Diagram ......................................................................... 40
Figure 3.8: User Side Activity Diagram ............................................................................ 41
Figure 4.1: Class Diagram ................................................................................................. 44
Figure 4.2: Entity Relationship Diagram ........................................................................... 45
Figure 4.3: Sequence Diagram of Enrollment ................................................................... 47
Figure 4.4: Sequence Diagram of Preprocessing ............................................................... 47
Figure 4.5: Sequence Diagram of Training ........................................................................ 48
Figure 4.6: Sequence Diagram of Testing ......................................................................... 48
Figure 5.1: Deployment Diagram ...................................................................................... 51

Table of Tables
Table 6.1: Main Menu Test Case ....................................................................................... 60
Table 6.2: Enrollment Test Case ........................................................................................ 60
Table 6.3:User Input Handling Test Case .......................................................................... 61
Table 6.4: Face Detection Test Case.................................................................................. 61
Table 6.5: Database Inserting Test Case ............................................................................ 62
Table 6.6: Training Test Case ............................................................................................ 62
Table 6.7: Testing Test Case .............................................................................................. 63
Chapter 1 Introduction
BCIS 1

CHAPTER 1
Introduction

BCIS 2

1. Introduction
The capacity of human being to recognize faces is quite remarkable. It is nearly automatic
and so intelligent that it persists even when the shape of a persons face changes with the
passage of time. We are supposed to bring this property of Natural Intelligence in
Artificial Intelligence so that we could recognize faces electronically. There are no built
in features in computers to perform this task, we have to program it.
There are many approaches used for the identification of person in which the most
common are Username and Password to login a Personal Computer, PIN (Person
Identification Number) Code systems. For example ATMs, Token systems such as
National ID Card number or vehicles license etc. But there were many problems with
these methods like losing card, theft, forgetting password, cards getting washed out and
many other problems like this. Due to these problems, there has been developed
considerable interest in biometric identification systems, which use pattern characters.
1.1. Problem Statement
Considering of an image representing a frame taken from video stream, automatic face
recognition is a particularly complex task that involves detection and location of faces in
a cluttered background followed by normalization and recognition. The human face is a
very challenging pattern to detect and recognize, because while its anatomy is rigid
enough so that all faces have the same structure, at the same time there are a lot of
environmental and personal factors affecting facial appearance. The main problem of face
recognition is large variability of the recorded images due to pose, illumination
conditions, facial expressions, use of cosmetics, different hairstyle, presence of glasses,
beard, etc. Images of the same individual taken at different times, may sometimes exhibit
more variability due to the aforementioned factors (intrapersonal variability), than images
of different individuals due to gender, race, age and individual variations (extra personal
variability). One way of coping with intrapersonal variations is including in the training
set images with such variations. And while this is a good practice for variations such as
facial expressions, use of cosmetics and presence of glasses or beard, it may not be
successful in case of illumination or pose variations.

BCIS 3

1.2. Purpose
This project is aimed to identify the persons. Here the technique is we already store some
images of the person in our database along with his details. These images are stored in
database record so to identify any person. If any image is matched then we predict that he
is only the required person. Thus using this project it provides a very friendly
environment for users to identify persons very easy.
1.3. Scope
The scope of the project is confined to store the image. When a person has to be identified
the images stored in the database are compared with the existing details.
1.4. Definition, Acronyms, and Abbreviations
BPIS: Biometric Person Identification System
1.5. Objective of the Project
The aim of this project is to come up with a simple and improved implementation for
person identification using personal computers. In this project, we have used for the
algorithms Eigen Faces and Principal Component Analysis (PCA) to determine
similarities between faces which consider different poses of a person and also to maintain
the profiles of the persons along with their faces.
1.6. Positioning
1.6.1. Problem Statement

The Problem
of

Manually Identify the persons.

Affects

Security agencies and investigation departments (employees).

The Impact of
which is

Administration needs to manually manage the record of the
persons and its hard and time consuming to identify the persons.

A successful
solution would
be

Proposed System will automatically manage the information of the
persons. And identify the persons by biometric techniques.
BCIS 4

1.7. Main Project Achievements
The project achievements can be stated as follows:
Adaptation of two algorithms in JAVA that compute similarity between faces.
PCA and Eigen Faces.
Insertion, Deletion and Updating of user profiles along with their faces.
Development of a Graphical User Interface which allows user to carry out
experiment to test the system.
A set of experiment carried out with 4 different pictures of 5 to 10 different
subjects and found the recognition result of the system to be 80%.
1.8. Product Overview
The general description comprises of following phases
1.8.1. Product Perspective
The product is a standalone application which is proposed in desktop based, and a
windows service for executing product. It integrates with MySQL on one end, and
interfaces with a GUI package to provide the visual content. In between, business logic is
involved.
1.8.2. Assumptions and Dependencies
Assuming that user must have Camera attached with the system whether it was a built-in
Webcam or separate USB Cam.
1.8.3. Cost and Pricing
To be decided.

BCIS 5

1.8.4. System Model
The following figure shows the system model.

1.9. Product Features
Following are the brief descriptions of the component of the system model.
1.9.1.1. Enrollment
In this process, users personal info is entered and stored in the data base. Users frontal
face samples are taken. Face samples are verified through Face Detection. User is
provided with an option to save a sample. At least four samples of a user are taken.
1.9.1.2. Training
Classes are computed and Eigen faces of classes are generated that were used as face
space for recognition process.
User Info
Enrollment
Face Samples
PCA
Eigen Vectors
Preprocessing
Eigen Values
Training
Recognition
BCIS 6

1.9.1.3. Recognition
Face sample is taken of the person to be recognized. Again the face sample is verified
through Face Detection. Face sample is preprocessed. Similarity scores are computed and
the best match is found.
1.9.2. User Characteristics
Running this software does not require any advance knowledge of computer. This
software is very user friendly so any person that has basic knowledge of computer can use
this software.
1.9.3. Functional Requirement
Following are some basic requirements are described briefly:
- Enrollment
Taking the snapshots of the user
Detecting face in the snapshot
Maintaining the profile
- Administration of user profile
Insert
Update
Delete
- Identification
Taking the snapshot
Detecting face
Recognizing face
1.9.4. Organization of Report
Chapter 2 contains the literature reviewed for the BPIS system. Chapter 3 explains the
basic concepts which include preprocessing, Object Detection and Face Recognition.
Chapter 4 explains Analysis and Design in BPIS system which has Use Cases, Use Case
Diagram, SSD (System Sequence Diagram), Domain Model and Class Diagram. Chapter
5 describes the Implementation Details; Chapter 6 describes the testing mechanisms.
Chapter 7 contains user manual. Chapter 8 and Chapter 9 tell us about challenges and
further issues in the development of project. Appendix A contains background
mathematics that is needed to understand the algorithms.
BCIS 7

1.9.5. Tools and Technologies
The main tools and technologies used in the development of BPIS are:
Eclipse IDE
SQL server
Web Camera
SPARX Enterprise Architect
Chapter 2 Literature Survey & Basic Concepts
BCIS 8

CHAPTER 2
Literature Survey
And
Basic Concepts

BCIS 9

2. Literature Survey
There is a lot of study done for the development of this project. There are many projects
developed in the area of biometrics and many techniques used for Face Recognition.
2.1. Projects Developed
2.1.1. ImageWare Software, Inc.
ImageWare Systems, Inc. (IWS) is a leader of cutting-edge, identity-based, credential
management solutions driven by biometric technology
Delivering next-generation biometrics as an interactive and Scalable cloud-based
solution
IWS brings together cloud and mobile technology to offer multi-factor authentication for
smart phone users and mobile clients. IWS supports multi-modal biometric authentication
including voice, fingerprint, and facial recognition. All of which can be combined
alongside other authentication and access control facilities including tokens, digital
certificates, passwords, and PINS, to provide the ultimate level of assurance and
accountability for corporate networks and Web applications from mobile phones and
tablets to PC desktop environment.
Key solution offerings:

Biometric Enginee 2.0
This is a scalable, agnostic multibiometric identity management solution that ensures only
valid individuals gain access to controlled areas or attains secure documents.
It is agnostic in biometric algorithm and hardware and can manage tens of millions of
individuals using standard off the-shelf hardware, minimizing the cost of ownership.

GoMobile interactive
It is a platform allowing any mobile application to be secured with biometric
identification. It enables a range of unprecedented activities from secure sharing of
sensitive information to the realization of a true mobile wallet.
The biometrics-based authentication gives users the confidence that their personal
information is secure, while the push marketing capabilities allow companies unparalleled
interactivity that can be personalized to the needs and interests of their customers.
BCIS 10

2.1.2. VeriLook SDK (Face identification for stand-alone or Web
applications)
NEUROtechnology has developed a face recognition algorithm. VeriLook facial
identification technology is designed for biometric systems developers and integrators.
The technology assures system performance and reliability with live face detection,
simultaneous multiple face recognition and fast face matching in 1-to-1 and 1-to-many
modes.
VeriLook is available as a software development kit that allows development of stand-
alone and Web-based solutions on Microsoft Windows, Linux, Mac OS X and Android
platforms.
VeriLook algorithm implements advanced face localization, enrollment and matching
using robust digital image processing algorithms with the same high recognition quality
on PCs, embedded and mobile devices. The algorithm is able to perform simultaneous
multiple face processing in a single frame, as well as live face detection and gender
classification. Facial feature points can be optionally extracted.
2.1.3. TrueFace
This software is developed at Miros, Inc. Companys software allows for face recognition
through use of a video camera and a computer; site provides free demo of products to
demonstrate how your face can be used to access secure web pages; included its
information on TrueFace, a biometric that eliminates cards for check cashing ATMs.
2.1.4. FaceSDK
FaceSDK is a software development kit that lets you add facial recognition
features to your desktop and web applications.
Compatible with 32- and 64-bit apps in multiple platforms (Windows, Mac and Linux)
FaceSDK enables you to use facial recognition technology in your software, so that you
can develop apps to authenticate users with webcams, find matching faces on multiple
images, recognize people in graphic editors or create morphing and animation effects.
BCIS 11

FaceSDK can be used both on online and desktop applications, anywhere where accurate
facial recognition is required from funny face animation effects to biometric
authentication for high-security systems.
This SDK includes several samples in different programming languages (Microsoft
Visual C++, C#, Visual Basic and more) for testing purposes, as well as a demo to test all
the SDKs possibilities right away.
2.1.5. Intelligent vision systems, Inc.
Product focus on face recognition (searches in data bases of faces), face verification
(verification of identity of a single individual), electronic look control (software activates
an electronic door lock mechanism via printer port and simple interface), and
ATM/secure access/computer security applications
2.1.6. FaceIt Pc
This software is developed at Visionics Corporation. Information on the companys
biometric recognition system (a software engine that automatically detects, locates and
identifies human faces from live video or static images) is provided; company focuses on
building applications for banking, Intranet security, surveillance, time & attendance, law
enforcement, national IDs, application security, multimedia search, passport control,
service kiosks, entitlement programs, voter registration, enterprise security, database
access, teleconferencing, airport/luggage security, border control, consumer research, and
vehicular sentry.

BCIS 12

2.2. Techniques for Face Recognition
Following are the techniques used for the purpose of Face Recognition system.
2.2.1. Geometric Features
Many people have explored geometric feature based methods for face recognition.
Kanade presented an automatic feature extraction method based on ratios of distances and
reported a recognition rate of 45-75% with data base of 20 people. Burnelli and Paggio
compute a set of geometrical features such as nose width and length, mouth position, and
chin shape. They report a 90% recognition rate on a database of 47 people. However they
show that a simple template matching scheme provides 100% recognition for the same
database. Cox et al. have recently introduced a mixture-distance technique which
achieves a recognition rate of 95% using a query database of 95 images from a total of
685 images. Each face is represented by 30 manually extracted distances.
Systems, which employee precisely measured distances between features, may be most
useful for finding possible matches in a large mug-shot database. For other applications
automatic identification of these points would be required, and the resulting system would
be dependent on the accuracy of the feature location algorithm.
2.2.2. Eigen Faces
Eigenfaces is the name given to a set of eigenvectors when they are used in the computer
vision problem of human face recognition. The approach of using eigenfaces for
recognition was developed by Sirovich and Kirby (1987) and used by Matthew Turk
and Alex Pentland in face classification. The eigenvectors are derived from the
covariance matrix of the probability distribution over the high-dimensional vector space
of face images. The eigenfaces themselves form a basis set of all images used to construct
the covariance matrix. This produces dimension reduction by allowing the smaller set of
basis images to represent the original training images. Classification can be achieved by
comparing how faces are represented by the basis set.
Turk and Pentland present results on a database of 16 subjects with various head
orientation, scaling and lighting. Their images appear identical otherwise with little
variation in facial expression, facial details, pose etc. For lighting, orientation, and scale
variation their system achieves 96%, 85% and 64% correct classification respectively.
Scale is normalized to the eigenface size based on an estimate of their head size. The
BCIS 13

middle of the faces is accentuated, reducing any negative affect of changing hairstyle and
backgrounds.
2.2.3. Graph Matching
An approach to face recognition is well-known method of Graph matching. Lades et al.
present dynamic link Architecture for distortion invariant object recognition, which
employs elastic graph matching to find the closest stored graph. Objects are presented
with spare graphs whose vertices are labeled with a multi-resolution description in terms
of a local power spectrum, and whose edges are labeled with geometric distances. They
present good results with a database of 87 people and test images composed of different
expressions and faces turned 15 degrees. The matching process is computationally
expensive, taking roughly 25 seconds to compare an image of 87 stored objects when
using a parallel machine with 23 transputers. Wishok et al. use an updated version of the
technique and compare 300 faces against 300 different faces of same people taken from
the FERET database. They report a recognition rate of 97.3%.
2.2.4. Fisher Faces (Linear Discriminate Analysis algorithm)
Linear Discriminate Analysis algorithm is also used in face recognition. Originally
developed in 1936 by R.A. Fisher, Discriminant Analysis is a classic method of
classification that has stood the test of time. Discriminant analysis often produces models
whose accuracy approaches (and occasionally exceeds) more complex modern methods.
LDA is similar to Principal Component Analysis (PCA). In both PCA and LDA we look
for the linear combinations of the variables which explain the given data well. LDA
explicitly attempts to model the difference between the classes of data. PCA on the other
hand does not take into account any difference in class but builds the feature
combinations based on differences rather than similarities.
In a computer based face recognition system, each face is represented by a large number
of pixel values. Linear discriminant analysis is primarily used here to reduce the number
of features to a more manageable number before classification. Each of the new
dimensions is a linear combination of pixel values, which form a template. The linear
combinations obtained using Fisher's linear discriminant are called Fisher Faces.
BCIS 14

2.2.5. Template Matching
A simple version of template matching is that a test image represented as a two
dimensional array of intensity values is compared using a suitable metric, such as the
Euclidean distance, with a single template representing the whole face. There are several
other more sophisticated versions of template matching on face recognition. One can use
more than one face template from different viewpoints to represent an individual's face. A
face from a single viewpoint can also be represented by a set of multiple distinctive
smaller templates. The face image of gray levels may also be properly processed before
matching. In, Bruneli and Poggio automatically selected a set of four features templates,
i.e., the eyes, nose, mouth, and the whole face, for all of the available faces. They
compared the performance of their geometrical matching algorithm and template
matching algorithm on the same database of faces, which contains 188 images of 47
individuals.
The template matching was superior in recognition (100 percent recognition rate) to
geometrical matching (90 percent recognition rate) and was also simpler. Since the
principal components (also known as eigenfaces or eigenfeatures) are linear combinations
of the templates in the data basis, the technique cannot achieve better results than
correlation, but it may be less computationally expensive. One drawback of template
matching is its computational complexity. Another problem lies in the description of
these templates. Since the recognition system has to be tolerant to certain discrepancies
between the template and the test image, this tolerance might average out the differences
that make individual faces unique. In general, template-based approaches compared to
feature matching are a more logical approach.
2.2.6. Neural Network Approach
The attractiveness of using neural network could be due to its non-linearity in the
network. Hence, the feature extraction step may be more efficient than the linear
Karhunen-LoeAve methods. One of the first artificial neural network (ANN) techniques
used for face recognition is a single layer adaptive network called WISARD which
contains a separate network for each stored individual. The way of constructing a neural
network structure is crucial for successful recognition. It is very much dependent on the
intended application.
BCIS 15

For face detection, multilayer perceptron and convolutional neural network have
been applied.
For face verification, Cresceptron multi resolution pyramid structure is used.
Lawrence et al. proposed a hybrid neural network, which combined local image sampling,
a self-organizing map (SOM) neural network, and a convolutional neural network. The
SOM provides a quantization of the image samples into a topological space where inputs
that are nearby in the original space are also nearby in the output space, thereby providing
dimension reduction and invariance to minor changes in the image sample. The
convolutional network extracts successively larger features in a hierarchical set of layers
and provides partial invariance to translation, rotation, scale, and deformation. The
authors reported 96.2 percent correct recognition on ORL database of 400 images of 40
individuals. The classification time is less than 0.5 second, but the training time is as long
as 4 hours. Lin et al used probabilistic decision-based neural network (PDBNN), which
inherited the modular structure from its predecessor, a decision based neural network
(DBNN).
The PDBNN can be applied effectively to
1) Face detector: This finds the location of a human face in a cluttered image,
2) Eye localizer: This determines the positions of both eyes in order to generate
meaningful feature vectors, and
3) Face recognizer: A hierarchical neural network structure with non-linear basis
functions and a competitive credit-assignment scheme.
PDBNN-based biometric identification system has the merits of both neural networks and
statistical approaches, and its distributed computing principle is relatively easy to
implement on parallel computer. In, it was reported that PDBNN face recognizer had the
capability of recognizing up to 200 people and could achieve up to 96 percent correct
recognition rate in approximately 1 second. However, when the number of persons
increases, the computing expense will become more demanding. In general, neural
network approaches encounter problems when the number of classes (i.e., individuals)
increases. Moreover, they are not suitable for a single model image recognition task
because multiple model images per person are necessary in order for training the systems
to optimal parameter setting.
BCIS 16

2.2.7. Stochastic Modeling
Stochastic modeling of non-stationary vector time series based on hidden Markov models
(HMM) has been very successful for speech applications. Samaria and Fallside applied
this method to human face recognition. Faces were intuitively divided into regions such
as the eyes, nose, mouth, etc., which can be associated with the states of a hidden Markov
model. Since HMMs require a one-dimensional observation sequence and images are two
dimensional, the images should be converted into either 1D temporal sequence or 1D
spatial sequence. In, a spatial observation sequence was extracted from a face image by
using a band sampling technique. Each face image was represented by a 1D vector series
of pixel observation. Each observation vector is a block of L lines and there is an M lines
overlap between successive observations. An unknown test image is first sampled to an
observation sequence. Then, it is matched against every HMMs in the model face
database (each HMM represents a different subject). The match with the highest
likelihood is considered the best match and the relevant model reveals the identity of the
test face. The recognition rate of HMM approach is 87 percent using ORL database
consisting of 400 images of 40 individuals. Pseudo 2D HMM was reported to achieve a
95 percent recognition rate in their preliminary experiments. Its classification time and
training time were not given (believed to be very expensive). The choice of parameters
had been based on subjective intuition.
2.2.8. N-tuple classifiers
Conventional n-tuple systems have the desirable features of super-fast single-pass
training, super-fast recognition, conceptual simplicity, straightforward hardware and
software implementations and accuracy that is often competitive with other more
complex, slower methods. Due to their attractive features, n-tuple methods have been the
subject of much research. In conventional n-tuple based image recognition systems, the
locations specified by each n-tuple are used to identify an address in a look-up-table. The
contents of this address either use a single bit to indicate whether or not this address was
accessed during training, or store a count of how many times that address occurred.
While the traditional n-tuple classifier deals with binary-valued input vectors, methods
using n-tuple systems with integer-valued inputs have also been developed. Allinson and
Kolcz have developed a method of mapping scalar attributes into bit strings based on a
combination of CMAC and Gray coding methods. This method has the property that for
BCIS 17

small differences in the arithmetic values of the attributes, the hamming distance between
the bit strings is equal to the arithmetic difference. For larger values of the arithmetic
distance, the hamming distance is guaranteed to be above a certain threshold.
The continuous n-tuple method also shares some similarity at the architectural level with
the single layer look-up perceptron of Tattersall et al, though they differ in the way the
class outputs are calculated, and in the training methods used to configure the contents of
the look-up tables (RAMS).
In summary, no existing technique is free from limitations. Further efforts are required to
improve the performances of face recognition techniques, especially in the wide range of
environments encountered in real world.

BCIS 18

2.3. Basic Concepts
Project has been divided into following modules/functions.
2.3.1. Preprocessing
All pictures from the face database (included the subject to be tested) are preprocessed
because the system uses intensity values of the images to train itself and test an image.
The preprocessing is performed in following steps:
2.3.1.1. Gray Scale levels
Converts pictures to 256 gray scale levels because the algorithm operates on intensity
values.
2.3.1.2. Geometric normalization
Taking into account the face annotations using OpenCV Object Detection algorithm for
face, the method resize, scale, rotate and reflect the picture if needed in order to make it
centered and symmetric with relation to the face.
2.3.2. Object Detection
Object detection (face detection particularly in our system) is a basic requirement of face
recognition systems. There are some techniques that extract certain features of the object
of interest from the image and then use heuristics to find combination of those features in
the image. It is hard to extract such features because sometimes object in the image is not
properly placed or there is shadow on half of the object or a half of the object is not
visible due to lighting conditions. Statistical model-based techniques are quite fair.
Statistical model is first trained by providing multiple positive samples and negative
samples and then used to detect objects.
Face detection is a computer technology that determines the locations and sizes of human
faces in digital images. Face tracking is an extension of face-detection when applied to a
video image. Some of the face-detection techniques are Feature-based methods, using
skin color to find face segments, detecting a blinking eye pattern in a video image and
template-based matching. In Feature based technique, the content of a given region of an
image is transformed into features, after which a classifier trained on example faces
decides whether that particular region of the image is a face, or not.
Intels OpenCV (Open Source Computer Vision) library uses such a statistical model to
detect objects. OpenCV provides low-level and high level APIs for face/object detection.
BCIS 19

A low level API allows users to check an individual location within the image by using
the classifier cascade to find whether it contains a face or not. Helper functions calculate
integral images and scale the cascade to a different face size (by scaling the coordinates
of all rectangles of Haar-like features) etc. Alternatively, the higher-level function if
cvDetectObjects does this all. Figure shows the algorithms results
.
2.3.3. Face Recognition
Following is the description of the technique used to recognize faces.
2.3.3.1. Introduction
We used EigenFaces for face recognition. This approach was first developed by Sirovich
and Kirby (1987) and used later by Matthew Turk and Alex Pentland.
A dataset, such as a digital image, consists of a large number of inter-related variables.
Using Principal Component Analysis, the dimensionality of the dataset is reduced while
retaining as much as variation in the dataset as possible. The datasets are transformed to a
new set of uncorrelated variables called the principal components. These principal
components are ordered in such a way that the first few retain most of the variation
present in all of the original variables.
PCA method is applied in face recognition by discriminating the image data into several
classes. There will be a lot of noise in the image caused by differing lighting conditions,
pose and so on. Despite these noises, there are patterns that can be observed in the image
such as the presence of eyes, mouth or nose in a face and the relative distances between
these objects. PCAs aim is to extract these patterns or the features from the image.
BCIS 20

In the domain of face recognition, the principal components (features) are referred to as
Eigen faces. The original image can be reconstructed by summing up all the Eigen faces
in the right proportion by adding more weights to the real features of the face. Certain
Eigen faces that do not contribute to the important face features are omitted. This is
necessary because of performance issues while doing large computations. This idea is
applied in our approach where Eigen faces are prepared from a set of training images and
stored first. Then Eigen face is prepared for the input test image and compare with the
training images. The matching image is the image having similar weights in the test
database.
2.3.3.2. Concept
Face recognition using Eigen faces approach was initially developed by Sirovich and
Kirby and later used by Matthew Turk and Alex Pentland. They showed that a collection
of face images can be approximately reconstructed by storing a small collection of
weights for each face and a small set of standard pictures.
Using Principal Component Analysis on a set of human face images, a set of Eigen faces
can be generated. Eigen faces are a set of eigenvectors used mostly in human face
recognition. Eigenvectors are a set of features that characterize the variation between
face images. These eigenvectors are derived from the covariance matrix of the probability
distribution of the high-dimensional vector space of faces of human beings. The main
idea here is to use only the best Eigen faces that account for the major variance within the
set of face images. By using lesser Eigen faces, computational efficiency and speed is
achieved. The Eigen faces are the basis vectors of the Eigen face decomposition.
Below are the steps of face recognition process:
A training set of same resolution digital images is initially prepared.
The images are stored as a matrix with each row corresponding to an image.
Each image is represented as a vector with (r X c) elements where r and c are
the number of rows and the number of columns respectively.
An average image is calculated from the individual training set images.
For each image, the deviation from the average image is then calculated and
stored.
The Eigen vectors and Eigen values are then calculated. These represent the
directions in which the training set images differ from the average image.
BCIS 21

A new image is then subtracted from the average image and projected into Eigen
face space.
This is compared with the projection vectors of training faces and the matching
image is determined.
A face image is represented by a two dimensional N by N array of intensity values or a
vector of dimension N2. If there is an image of size 128 by 128, then that can be said as a
vector of dimension 16384. Or, this is equivalent to one point in a 16384-dimensional
space. A group of images then maps to a collection of points in this image space. The
training images chosen are all of same dimensions. We need to find the vectors that best
represent the distribution of face images within this image space and these vectors that
define the sub-space of face images are termed as face space. Each vector of length N2
represents an image of dimension N by N and is a linear combination of the original face
images. These vectors are termed as Eigen faces because these are the vectors of the
covariance matrix corresponding to the original face images and they have face-like
appearance.
Flow chart of the Algorithm is shown in Figure.2.1.
BCIS 22

Start training set image
preparation with the same
resolution images
Store image as a matrix represented
as a vector with (r X c) elements
Calculate average image from training
images
Calculate and store deviation of each
image from the average image
Calculate Eigenvectors and Eigenvalues
of the covariance matrix
Calculate covariance matrix
Store the
principal
components
End training set
preparation
Test
Input
Image
Subract from the
average image
Project into
Eigenface space
Compare
with
training
vectors
Determine the minimum
Euclidean distance which is
the matching image
End face
recognition
Training set preparation Face recognition stage

Figure 2.1: Flowchart of Face Recognition using Eigen Faces

2.3.3.3. Principal Component Analysis (PCA)
It is a way of identifying patterns in data, and expressing the data in such a way as to
highlight their similarities and differences. Since patterns in data can be hard to find in
data of high dimension, where the luxury of graphical representation is not available,
PCA is a powerful tool for analyzing data. The other main advantage of PCA is that once
you have found these patterns in the data, and you compress the data, i.e. by reducing the
number of dimensions, without much loss of information.
BCIS 23

PCA is technic widely used with regard to face recognition. The principle on which it
works is that face pictures have common objects (nose, eyes, mouth, etc). Every picture
from the face database is treated as pixel vector. The algorithm computes the covariance
matrix with all these vectors and extracts the Eigen vectors and corresponding Eigen
values. Figure illustrates the Eigen value spectrum of face data. PCA algorithm retains
only a few eigenvectors to form a new basis (Eigen Basis).
PCA Training
The training is the common training performed with particular settings. This is the part of
the process and computes the covariance matrix and the Eigen basis. Following are the
PCA steps:
I is an N2x1 vector, corresponding to an NxN face images I.
The idea is to represent I (u=I - mean face) into a low-dimensional space:
u mean = w1u1 + w2u2 + . . . wK uK (K<<N2)
Computation of the Eigen faces
Step 1: obtain face images I1, I2, ..., IM (training faces)
(very important: the face images must be centered and of the same size)

Step 2: represent every image Ii as a vector Ii
Step 3: compute the average face vector +:
BCIS 24

Step 4: subtract the mean face:
ui = Ii +
Step 5: compute the covariance matrix C:

Step 6: compute the eigen vectors
i
of AA
T
The matrix AA
T
is very large ----> not practical!!
Step 6.1: consider the matrix A
T
A (MxM matrix)
Step 6.2: compute the eigen vectors v
i
of A
T
A
A
T
Av
i
=
i
v
i

What is the relationship between us
i
and v
i
?

Thus, AA
T
and A
T
A have the same eigenvalues and their eigen vectors are related
as follows:
i=
Av
i
Note 1: AA
T
can have up to N
2
eigenvalues and eigenvectors.
BCIS 25

Note 2: A
T
A can have up to M eigenvalues and eigenvectors.
Note 3: The M eigenvalues of A
T
A (along with their corresponding
eigenvectors) correspond to the M largest eigenvalues of AA
T
(along with
their corresponding eigenvectors).
Step 6.3: compute the M best eigenvectors of AA
T
:
i=
Av
i
(important: normalize
i
such that ||

i
||=1)
Step 7: Keep only K eigenvectors (corresponding to the K largest eigenvalues)
Representing faces into the basis
Each face (minus the mean) in the training set can be represented as a linear
combination of the best K eigenvectors:

(We call the
j
s Eigen faces)

Each normalized training face u
i
is represented in this basis by a vector:

2.3.3.4. Face Recognition Using Eigenfaces
Given an unknown face image I (centered and of the same size like the training faces)
follow these steps:
Step 1: normalize I: u = I +
BCIS 26

Step 2: project on the eigenspace

Step 3: represent u as:

Step 4: find er = min
l
||OO
l|
|
Step 5: if er < Tr, then I is recognized as face l from the training set.
The distance e
r
is called distance within the face space (difs)
Comment: we can use the common Euclidean distance to compute er, however, it has
been reported that the Mahalanobis distance performs better:

(variations along all axes are treated as equally significant)
2.3.3.5. Face Detection Using Eigenfaces
Given an unknown image I
Step 1: compute u = I +
Step 2: compute

BCIS 27

Step 3: compute e
d
= ||u u||
Step 4: if e
d
< Td, then I is a face.
The distance e
d
is called distance from face space (dffs)

Chapter 3 System Analysis
BCIS 28

CHAPTER 3
System Analysis

BCIS 29

3. Analysis
Requirements are capabilities and conditions to which the system and more broadly the
project must conform. A prime challenge of requirements work is to find, communicate and
remember (that usually means record) what is really needed, in a form that clearly speaks to
client and development team members.
3.1. Use Cases Model
3.1.1. Use Case Diagram
A use case is a sequence of actions that provide a measurable value to an actor. Another way
to look at it is a use case describes a way in which a real-world actor interacts with the
system. In a system use case you include high-level implementation decisions. System use
cases can be written in both an informal manner and a formal manner. Use case Diagrams are
created to visualize the relationship between actors and use cases. A use case diagram is also
used to capture the system functionality as seen by the user.

Figure 3.1: Use Case Diagram

BCIS 30

3.1.2. Use Cases Descriptions in Brief
In software engineering, a use case diagram in the Unified Modeling Language (UML) is a
type of behavioral diagram defined by and created from a Use-case analysis. Its purpose is to
present a graphical overview of the functionality provided by a system in terms of actors,
their goals (represented as use cases), and any dependencies between those use cases.
The main purpose of a use case diagram is to show what system functions are performed for
which actor. Roles of the actors in the system can be depicted.
Use Case diagrams are formally included in two modeling languages defined by the OMG.
Both the UML and SysML standards define a graphical notation for modeling use cases with
diagrams. One complaint about the standards has been that they do not define a format for
describing these use cases. Generally, both graphical notation and descriptions are important
as they document the use case, showing the purpose for which an actor uses a system.
The use case diagram shows the position or context of the use case among other use cases. As
an organizing mechanism, a set of consistent, coherent use cases promotes a useful picture of
system behavior, a common understanding between the customer/owner/user and the
development team.

BCIS 31

3.1.3. Use Case Description in Detail
Following are the cases derived during Analysis.
3.1.3.1. Use Case UC-01: Enrollment
Use Case ID UC-01
Name Enrollment
Primary Actor Operator / Administrator
Secondary Actor User
Stakeholders and
Interests
Operator: Wants accurate and proper face samples of the person i.e.
user.
User: Wants his enrollment into the system.
Preconditions Operator initiated the system. Also camera is attached to the
system.
Post Conditions Persons face is detected, his accurate face samples captured and
stored and his personal information is also entered.
Main Success
Scenario
1. Operator enters users personal information.
2. User sits in front of the camera.
3. User is required to stay still and look directly towards the
camera.
4. System detects users face in the snapshot of the user.
5. Save the face sample if the face is detected accurately.
6. Discard if face is not detected accurately. Repeat steps 4-6
times at least 4 times.
7. Users face samples are stored on the disk.
Scenario Extensions 1a. At any time, system fails:
1. Operator restarts System.
2a. Camera is attached but it doesnt display the video.
1. Make sure device driver is installed
2. Make sure any other application is not using the
camera.
3a. Face is not going to be detected.
1. Lighting conditions should be sharp
2. User is required to stay still and towards the camera.
3. There are not two faces in front of the camera.
BCIS 32

3.1.3.2. Use Case UC-02: Preprocessing
Use Case ID UC-02
Name Preprocessing
Primary Actor Operator/ Administrator
Stakeholders and
Interests
Operator: face samples of all the enrolled users are preprocessed for
training of face recognition.
Preconditions Operator initiated the system. The face database already contains
the images of all the employees.
Post Conditions All the face samples enrolled are converted to gray scale, aligned
and scaled to be standard size.
Main Success
Scenario
1. System reads the image from the face database.
2. Convert it to gray scale.
3. Aligns with respect to face.
4. Scales to a standard size.
5. The image is cropped using the mask or cropped function
such that only the face from fore head to chin and cheek is
visible.
Repeat steps 4-6 for all the images in face data base
6. All the images are preprocessed.
To support recovery and ensure that database has not been
damaged.
2a. Preprocessed images are not accurate.
2. Enroll the user again
3. User is required to stay still and look directly
towards the camera

BCIS 33

3.1.3.3. Use Case UC-03: Training
Use Case ID UC-03
Name Training
Stakeholders and
Interests
Operator: Wants accurate and proper training of the system.
Preconditions Operator initiated the system. The face database already contains
the preprocessed images of all the employees.
Post Conditions All the calculations are performed. PCA is applied to the images,
eigenvectors and eigenvalues are calculated and system is ready for
recognition.
Main Success
Scenario
1. Different images are calculated.
2. Preprocessing is applied.
3. PCA is applied on the training images.
4. Eigenvectors and eigenvalues are generated.
5. Euclidean distance is applied.
6. System is ready for recognition.
damaged.
2a. A user is deleted from the database.
1. Delete his/her face samples from the database as
well.
2. Restart training.

BCIS 34

3.1.3.4. Use Case UC-04: Testing
Use Case ID UC-04
Name Testing
Stakeholders and
Interests
Operator: Wants accurate and proper training of the system.
User: Wants his recognition by a system.
Preconditions Training is performed, eigenvectors and eigenvalues are generated.
Camera is attached with the computer.
Post Conditions User is recognized and his personal information is returned. If not
recognized, proper message is displayed.
Main Success
Scenario
1. User sits in front of the camera.
2. User is required to stay still and look directly towards the
camera.
3. System detects users face in the snapshot of the user.
4. System preprocesses the face sample.
5. System starts matching the face sample with the enrolled
ones.
6. If matched, users personal information is returned.
7. If not matched, not found message is displayed.
8. May repeat steps 1-7 until the user is recognized.
damaged.
2a. A user is deleted from the database.
1. Delete his face samples from the database as well.
2. Restart training.
3a. Camera is attached but it doesnt display the video.
1. Make sure device driver is installed.
2. Make sure any other application is not using the
camera.
4a. Face is not going to be detected.
2. User is required to stay still and look directly
towards the camera.

BCIS 35

3.1.4. System Sequence Diagrams
A sequence diagram maps a scenario described by a use case in step by step detail to define
how objects collaborate to achieve your applications goals.
A lifeline in a sequence diagram represents an object and shows all its points of interaction
with other objects in events that are important to it. Lifelines start at the top of a sequence
diagram and descend vertically to indicate the passage of time. Interactions between objects,
messages and replies, are drawn as horizontal direction arrows connecting lifelines. In
addition, boxes known as combine fragments are drawn around sets of arrows to mark
alternative actions, loops, and other control structures.
UML sequence diagrams are typically used to describe object-oriented software systems, they
are also extremely useful as system engineering tools to design system architectures, in
business process engineering as process flow diagrams, as message sequence charts and call
flows for telecom/wireless system design, and for protocol stack design and analysis. The
following are the sequence diagram elements;
Actor
Represents an external person or entity that interacts with the system

Top Package::Actor

Object
Represents an object in the system or one of its components
object

Separator
BCIS 36

Represents an interface or boundary between subsystems, components or units (e.g., air
interface, Internet, network).

Call Message
A call (procedure) message between header elements
Call Message

Return Message
A return message between header elements
Return Message

Free Note
Documentation note that is free-flowing and can be placed anywhere in the diagram
{}

Following are the system sequence diagram of the corresponding use cases.
BCIS 37

3.1.4.1. SSD1: Enrollment

Figure 3.2: SSD1 Enrollment
3.1.4.2. SSD2: Preprocessing

Figure 3.3: SSD2 Preprocessing

BCIS 38

3.1.4.3. SSD3: Training

Figure 3.4: SSD3 Training
3.1.4.4. SSD4: Testing

Figure 3.5: SSD4 Testing

BCIS 39

3.2. Domain / Conceptual Model
A Domain Model is a high-level conceptual model, defining physical and abstract objects in
an area of interest to the Project. It can be used to document relationships between and
responsibilities of conceptual classes (that is, classes that represent the concept of a group of
things rather than Classes that define a programming object). It is also useful for defining the
terms of a domain.

Figure 3.6: Domain Model

BCIS 40

3.3. Activity Diagram
An activity diagram is a simple and intuitive illustration of what happens in a workflow, what
activities can be done in parallel, and whether there are alternative paths through the
workflow. Activity diagrams as defined in the Unified Modeling Language are derived from
various techniques to visually illustrate workflows.
In the Unified Modeling Language, activity diagrams can be used to describe the business
and operational step-by-step workflows of components in a system. n SysML the activity
diagram has been extended to indicate flows among steps that convey physical element (e.g.,
gasoline) or energy (e.g., torque, pressure).

3.3.1. Admin Side Activity Diagram

Figure 3.7: Admin Side Activity Diagram

BCIS 41

3.3.2. User Side Activity Diagram

Figure 3.8: User Side Activity Diagram

Chapter 4 System Design
BCIS 42

CHAPTER 4
System Design
BCIS 43

4. System Design
Software design sits at the technical kernel of the software engineering process and is
applied regardless of the development paradigm and area of application. Design is the
first step in the development phase for any engineered product or system. The designers
goal is to produce a model or representation of an entity that will later be built.
Beginning, once system requirement have been specified and analyzed, system design is
the first of the three technical activities -design, code and test that is required to build and
verify software.
The importance can be stated with a single word Quality. Design is the place where
quality is fostered in software development. Design provides us with representations of
software that can assess for quality. Design is the only way that we can accurately
translate a customers view into a finished software product or system. Software design
serves as a foundation for all the software engineering steps that follow. Without a strong
design we risk building an unstable system one that will be difficult to test, one whose
quality cannot be assessed until the last stage.
During design, progressive refinement of data structure, program structure, and
procedural details are developed reviewed and documented. System design can be viewed
from either technical or project management perspective. From the technical point of
view, design is comprised of four activities architectural design, data structure design,
interface design and procedural design. A good design must contain the following
characteristics;
In design phase of the software, alternative approaches must be considered based
on the requirements
Design should be traceable to the analysis model
Design should exhibit uniformity and integration
Design should be structured to accommodate any change
Design should be reviewed to minimize errors
Ensure the accurate translation of the requirements.
It must be readable and understandable.
It should provide complete picture of the software.
Design forms basis for programming and maintenance
BCIS 44

The Designing of system is an important stage during system development. The following
components are including in system design phase that are related to this project.
4.1. Class Diagram
Class or structural diagrams define the basic building blocks of a model. They are used
for static object modeling, describing what attributes and behavior a class has.

Figure 4.1: Class Diagram

BCIS 45

4.2. Entity Relationship Diagram
An entity-relationship model is an abstract and conceptual representation of data. Entity-
relationship modeling is a database modeling method, used to produce a type of
conceptual schema or semantic data model of a system, often a relational database, and its
requirements in a top-down fashion. Diagrams created by this process are called Entity-
Relationship Diagrams

Figure 4.2: Entity Relationship Diagram

BCIS 46

4.3. Sequence Diagrams
A sequence diagram is a kind of interaction diagram that shows how processes operate
with one another and in what order. It is a construct of a Message Sequence Chart. A
sequence diagram shows object interactions arranged in time sequence. It depicts the
objects and classes involved in the scenario and the sequence of messages exchanged
between the objects needed to carry out the functionality of the scenario. Sequence
diagrams are typically associated with use case realizations in the Logical View of the
system under development. Sequence diagrams are sometimes called event diagrams,
event scenarios
A sequence diagram shows, as parallel vertical lines (lifelines), different processes or
objects that live simultaneously, and, as horizontal arrows, the messages exchanged
between them, in the order in which they occur. This allows the specification of simple
runtime scenarios in a graphical manner.
UML sequence diagrams are typically used to describe object-oriented software systems,
they are also extremely useful as system engineering tools to design system architectures,
in business process engineering as process flow diagrams, as message sequence charts
and call flows for telecom/wireless system design, and for protocol stack design and
analysis. The following are the sequence diagram elements;

BCIS 47

4.3.1. SD1: Enrollment

Figure 4.3: Sequence Diagram of Enrollment
4.3.2. SD2: Preprocessing

Figure 4.4: Sequence Diagram of Preprocessing

BCIS 48

4.3.3. SD3: Training

Figure 4.5: Sequence Diagram of Training
4.3.4. SD4: Testing

Figure 4.6: Sequence Diagram of Testing
Chapter 5 Implementation
BCIS 49

CHAPTER 5
Implementation

BCIS 50

5. Implementation
This chapter is about the overview of the technologies used to develop this project.
Implementation is the important phase of software development life cycle where the
thoughts and ideas are given a physical existence. It is just like making your dreams true.
It is the time when software development progress in a full swing manner. An application
is a result of a successful implementation.
Implementation (software) perspective describes software implementations in a particular
technology (such as Java). In the UP, Implementation means programming and building
the system, not deploying it.
During implementation, developers translate the object model in to source code. This
includes implementing the attributes and methods of each object and integrating all the
objects such that they function as a single system.
The implementation activity span the gap between the detailed object design model and a
complete set of source code files that can be compiled together.
There are three types of Implementation:
Implementation of a computer system to replace a manual system. The problems
encountered are converting files, training users, and verifying printouts for
integrity.
Implementation of a new computer system to replace an existing one. This is
usually a difficult conversion. If not properly planned there can be many
problems.
Implementation of a modified application to replace an existing one using the
same computer. This type of conversion is relatively easy to handle, provided
there are no major changes in the files.
Implementation in Generic tool project is done in all modules. In the first module User
level identification is done. In this module every user is identified whether they are
genuine one or not to access the database and also generates the session for the user.
Illegal use of any form is strictly avoided.

BCIS 51

5.1. Tools and Technology
Following are the tools and technologies used to develop this project:
Eclipse IDE for JAVA Developers
SQL server
Web Camera
SPARX Enterprise Architect 7.5
Visual Paradigm for UML 11
5.2. Deployment Diagram
A deployment diagram models the run-time architecture of a system. It shows the
configuration of the hardware elements (nodes) and shows how software elements and
artifacts are mapped onto those nodes.

Figure 5.1: Deployment Diagram
Chapter 6 Testing
BCIS 52

CHAPTER 6
Testing

Chapter 6 Testing
BCIS 53

6. Software Testing
Testing is the last phase of the software development life cycle. Software testing is any
activity aimed at evaluating the capability of a system or software and determining that it
meets its required results and executing of the software with the intent of finding any
errors or shortcomings in it. Software testing is quite a difficult and complex process and
difficulty in software testing increases according to the complexity of software. Testing is
more than just debugging. The purpose of testing can be quality assurance, verification
and validation, or reliability estimation. Testing can be used as a generic metric as well.
Software testing is a trade-off between budget, time and quality.
Software is not unlike other physical processes where inputs are received and outputs are
produced. Where software differs is in the manner in which it fails. Most physical
systems fail in a fixed and reasonably small set of ways. By contrast, software can fail in
many bizarre ways. Detecting all of the different failure modes for software is generally
infeasible.
Most of the defects in software are design errors, not manufacturing defects. These design
defects or software bugs will almost always exist in any software module with moderate
size: not because programmers are careless or irresponsible, but because the complexity
of software is generally intractable -- and humans have only limited ability to manage
complexity. It is also true that for any complex systems, design defects can never be
completely ruled out.
During the testing phase all the activities that occur are documented. There are various
reasons why clear documentation is required during testing. It helps the development
team to understand the bugs and fix them quickly. In case there is a change in the testing
team it will help the new team members to quickly understand the process and help in a
quick transition. The overall summary report of the testing process helps the entire project
team to understand the initial flaws in design and development and ensure that the same
errors are not repeated again.
Testing has an important place in SDLC (Software Development Life Cycle). The SQA
(Software Quality Assurance) team should be involved in the early phases of the project.
This will make them aware of the requirements and techniques that are being followed
enabling them to develop an appropriate test strategy.
Chapter 6 Testing
BCIS 54

The marketing group will bring the customer requirement.
The engineering group will access the feasibility.
The engineering and SQA groups will work together during the design phase and
generate the product specification.
The developers will start implementing while the SQA team is developing the test
plans and so on.
While the developers are ready with the code, the SQA team will be ready for
testing. The testing phase is no longer a bottleneck for product delivery because
the SQA team has been involved since the start of the project.
System testing is an empirical investigation conducted to provide stakeholders
with information about the quality of the product or service under test. Software
Testing also provides an objective, independent view of the software to allow the
business to appreciate and understand the risks at implementation of the software.
Test techniques include, but are not limited to, the process of executing a program
or application with the intent of finding software bugs
Chapter 6 Testing
BCIS 55

6.1. Why is Software Tested?
Regardless of the limitations, testing is an integral part in software development. It is broadly
deployed in every phase in the software development cycle. Typically, more than 50%
percent of the development time is spent in testing. Testing is usually performed for the
following purposes:
6.1.1. To Improve Quality:
Quality means the correctness to the specified design requirement. Being correct, the
minimum requirement of quality, means performing as required under specified
circumstances. Debugging, a narrow view of software testing, is performed heavily to find
out design defects by the programmer. Finding the problems and get them fixed is the
purpose of debugging in programming phase.
6.1.2. For Verification and Validation (V&V)
Another important purpose of testing is verification and validation (V&V). Verification is the
process of evaluating a system or component to determine whether the products of a given
development phase satisfy the conditions imposed at the start of that phase. Validation is the
process of evaluating a system or component during or at the end of the development process
to determine whether it satisfies specified requirements.
6.1.3. For Reliability Estimation
Software reliability has important relations with many aspects of software including the
structure and the amount of testing it has been subjected to. Based on an operational profile
testing can serve as a statistical sampling method to gain failure data for reliability
estimation.

Chapter 6 Testing
BCIS 56

6.2. Software Testing Techniques
There are many ways to conduct software testing, but the most common methods rely on
following steps:
6.2.1. Test Case Design
Test cases should be designed in such a way as to uncover quickly and easily as many errors
as possible. They should check the program by using and producing inputs and outputs that
are both correct and incorrect.
6.2.2. White Box Testing

In white box testing the tester is concerned with the internal coding of the software. In
this strategy the code is examined without being executed. White box testing is concerned
only with testing the software product; it cannot guarantee that the complete specification has
been implemented. Black box testing is concerned only with testing the specification; it
cannot guarantee that all parts of the implementation have been tested. White box testing can
indicate test considerations, which are not produced by black box testing. The converse is
also true; black box testing can produce test considerations, which are not produced by white
box testing.
6.2.3. Basic Path Testing
A path is a sequence of operations that runs from the start of the program to an exit
point. This is also called an end-to-end path. Path testing is concerned with testing paths of
the software.
Basic path testing is a white-box testing technique. It allows the design and the definition of a
basic set of execution paths. The test cases created from the basis set allow the program to be
executed in such a way as to examine each possible path through the program by executing
each statement at least one.
There are three converge criteria in path testing. Line converge requires execution of every
line of code at least once. Branch converge requires testing of all lines and all branches.
Condition converge requires each ways of the condition can be made true or false.
6.2.4. Regression Testing
It is very important testing technique. In this testing technique the old tests are tested
again. This can be performed after black box or white box testing.
Chapter 6 Testing
BCIS 57

6.2.5. Control Structure Testing
Condition testing can be utilized to design the test case which examines the logical conditions
in a program. It focuses on all conditions in the program and includes testing of both
relational expressions and arithmetic expressions.
6.2.6. Black Box Testing
Black-box testing on the other hand focuses on the overall functionality of the
software. This technique does not concern with the internal coding. That is why it is the
chosen method for designing test case use in functional testing. This method allows the
functional testing to uncover faults like incorrect or missing functions, errors in any of the
interfaces, errors in data structures or data bases and errors related to performance and
program initialization or termination. The software is executed without examining the code.
The tester acts as an end user and tries to find bugs.
6.3. Software Testing Strategies
In order to conduct a proper and through set of tests, the types of testing mentioned below
should be performed in the order in which they are described.
6.3.1. Unit Testing
Unit testing focuses verification effort on the smallest unit of application i.e. write notes.
Using the detailed design and the process specifications, testing is done to uncover errors
within the boundary of the module. All modules must be successful in the unit test before the
start of the integration testing begins. In this project each service can be thought of a module.
There are modules like schedule, budget, plans, to-do list, calculator, Giving different sets of
inputs has tested each module. When developing the module as well as finishing the
development so that each module works without any error. The inputs are validated when
accepting from the user. In this application developer tests the programs up as system.
Software units in a system are the modules and routines that are assembled and integrated to
form a specific function. Unit testing is first done on modules, independent of one another to
locate errors. This enables to detect errors. Errors resulting from integration between modules
are initially avoided.
6.3.2. Integration Testing
After the unit testing we have to perform integration testing. The goal here is to see if
modules can be integrated properly. The emphasis being on testing interfaces between
modules. This testing activity can be considered as testing the design and hence the emphasis
on testing module interactions. In this project integrating all the modules forms the main
Chapter 6 Testing
BCIS 58

system. When integrating all the modules we have checked whether the integration effects
working of any of the services by giving different combinations of inputs with which the two
services run perfectly before integration.
6.3.3. Function Testing
Function testing is a testing process that is a black-box in nature. It is aimed at examining the
overall functionality of the product. It usually includes testing of all the interfaces and should
therefore involve the clients in the process.
6.3.4. System Testing
System Testing tests all components and modules that are new, changed, affected by a
change, or needed to form the complete application. The system test may require involvement
of other systems but this should be minimized as much as possible to reduce the risk of
externally-induced problems. The emphasis in system testing is validating and verifying the
functional design specification and seeing how all the modules work together.
6.3.5. Acceptance Testing
Acceptance test is performed with realistic data of the client to demonstrate that the software
is working satisfactorily. Testing here is focused on external behavior of the system. The
internal logic of program is not emphasized. Test cases should be selected so that the largest
number of attributes of an equivalence class is exercised at once. The testing phase is an
important part of software development. It is the process of finding errors and missing
operations and also completes verification to determine whether the objectives are met and
the user requirements are satisfied.
6.4. BPIS Testing
It was kept in consideration that testing is to be done at all stages from analysis to the
implementation phase. Hence at analysis phase it was made sure that the requirements are
complete, correct and achievable. Similarly, at design phase it was kept in mind that the
design was complete and implemented. Keeping in view the importance of each testing
methods and strategy it was tried to make the software undergo through all types of tests.
6.5. Test case
A test case is a set of data items that the system processes as normal input. Test case contains
the required data for testing a piece of software. Good testing involves much more than just
running the program a few times to see whether it works. A successful test is the one that
finds error. For one use-case there can be more than one test case. Hence the test cases were
Chapter 6 Testing
BCIS 59

designed in such a way that somehow all the strategies and methods were covered at one
point or the other.
6.6. Deriving Test Cases

The steps for deriving test cases are:
- Using design or code, draw the corresponding flow graph.
- Determine the cyclamate complexity of flow graph.
- Determine a basic set of independent paths.
- Prepare test cases that will force execution of each path in the basic set.

Chapter 6 Testing
BCIS 60

6.7. Test Case Description
Test cases are described below.
6.7.1. Loading of Menu
Test Case ID TC-001 Test Engineer Abdul Bari, Asif Sharif
Use Case Reference UCRS1- Enrollment
Objective Verify that Main Menu appears when application is started for the
first time.
Product BPIS
Module Main
Environment Windows 7 Ultimate
Method:
Launch BPIS for the system.
A main menu should appear.
Verify that Main Menu appears when application is started for the first time after installation.
Comments Passed
Table 6.1: Main Menu Test Case
6.7.2. Enrollment Test Case
Objective Verify that Enrollment wizard stats when enrollment button is
clicked.
Product BPIS
Module Enrollment
Method:
Run the application.
A form will be opened having input text boxes.
Verify that Form is performing its functionality having enrollment page.
Comments Passed
Table 6.2: Enrollment Test Case

Chapter 6 Testing
BCIS 61

6.7.3. User Input handling
Objective Verify that user get error messages on wrong entries.
Product BPIS
Module Main GUI Handling
Method:
Main Form is Running.
Provide Wrong Input in Appropriate fields.
Verify that System Show Errors at illegal Input.
Comments Passed
Table 6.3:User Input Handling Test Case
6.7.4. Face Detection
Objective Verify that video of the user appears and face is detected.
Product BPIS
Module Face Detection
Method:
Start the application.
Verify that The System displays the video of the user and his face is properly detected.
Comments Passed
Table 6.4: Face Detection Test Case

Chapter 6 Testing
BCIS 62

6.7.5. Database Inserting
Objective Verify that new user record is inserted in database.
Product BPIS
Module Insert Person Information
Method:
Input the required information and face samples to the system.
Verify record has been saved into database.
Comments Passed
Table 6.5: Database Inserting Test Case
6.7.6. Training
Use Case Reference UCRS3- Training
Objective Verify that training starts when training button is clicked and
completes within time.
Product BPIS
Module Training
Method:
Click on training, verify that training wizard is started and completed within time.
Comments Passed
Table 6.6: Training Test Case

Chapter 6 Testing
BCIS 63

6.7.7. Testing
Use Case Reference UCRS4- Testing
Objective Verify that recognition starts when recognition button is clicked.
Product BPIS
Module Recognition
Method:
Verify that the face recognition test wizard is started.
Comments Passed
Table 6.7: Testing Test Case
Chapter 7 Conclusion
BCIS 64

CHAPTER 7
Conclusion

Chapter 7 Conclusion
BCIS 65

7. Conclusion
This project demonstrated the successful implementation of the following tasks:
(a) Face recognition using Eigen faces.
(b) Face Detection using Haar-like features.
Test results of face recognition using Eigen faces are shown below:
Statistic Results
Eigen faces
Number of images in Eigen faces database 16
Number of test images 30
Number of images successfully matched 28
Percentage of images successfully matched 93.33%

Future enhancements for face recognition can be expanded to handle more than 16 images in
the training database. Right now, adjusting the given input images or test images to the same
resolution is a manual process. The program can be expanded to adjust the image resolution
automatically without manual involvement
Chapter 8 User Manual
BCIS 66

Appendices
BCIS 72

APPENDICES

Appendices
BCIS 73

8. User Manual
Pc Minimum System Requirements
- OS : Windows XP , Windows 7
- CPU : Core 2 Duo E4500 2.5GHz or better
- RAM: 2GB
- HDD: 100MB of free space
8.1. Main Window
When the project is executed the opening window will look like the following bitmap image.
It contains some standard buttons for further processing.

Appendices
BCIS 74

8.2. Enrollment Wizard Window 1
Acquire the person face samples.

Appendices
BCIS 75

8.3. Enrollment Wizard Window 2
Input the criminals personal information and store it in database.

Appendices
BCIS 76

8.4. Recognition Window
This window is opened when the recognition of the person is required.

Bibliography and References
BCIS 77


BCIS 78

References
[1] M. A. Turk and A. P. Pentland, Face Recognition Using Eigen faces, In Proc. of
Computer Vision and Pattern Recognition, pages 586-591. IEEE, June 1991.
[2] A tutorial on Principal Components Analysis by Lindsay I Smith
[3] Jolliffe, I.T., Principal Component Analysis Springer Series in Statistics, 1-9, DOI:
10.1007/0-387-22440-8_1, 2002.
[4] Colt is a set of Open Source Libraries for High Performance Scientific and Technical
Computing written in Java
[5] Open source computer vision library http://opencv.org/
[6] Xiao-yan HU, An algorithm of Face Detection Based on Skin Color, Master's
thesis, Sichuan University, 2004.
[7] Peter N. Belhumeur, Joa P. Hespanha, and David J. Kriegman, Eigen faces vs.
Fisher faces: Recognition Using Class Specific Linear Projection, IEEE
TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,
VOL. 19, NO. 7, JULY 1997.
[8] Paul Viola and Michael J. Jones, Rapid Object Detection using a Boosted Cascade
of Simple Features, IEEE CVPR, 2001.

System Requirement Specification

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

System Requirement Specification

Uploaded by

Copyright:

Available Formats

BioMetric Criminals Identification System

You might also like