Fingerprint Database for Exam Authentication

TABLE OF CONTENTS
List of Tables
.2
List of Figures
..3
Abstract
.4
Introduction
.5
Literature Review ..
..9
Project Implementation
.20
Results and Discussions
36
Conclusion
.44
References
...
.
46
Appendix A
48
Appendix B
..55
LIST OF TABLES
Page
Table 1.1
Table 1.2
Normal Forms
Course Table
27
Table 1.3
Student Table
28
Table 1.4
Registration/Exam Enrollment Table
28
Table 1.5
Minutiae/Finger Details Table
29
2
Table 1.6
Course Table (Final)
29
Table 1.7
Bifurcations Table (Final)
36
Table 1.8
Terminations Table (Final)
37
Table 1.9
Students Table (Final)
Table 2.0
Finger Table (Final)
38
39
Table 2.1
Registration Table (Final)
40
LIST OF FIGURES
Page
Figure 2.0
9
A fingerprint image acquired by an Optical Sensor

3
Figure 2.2.1.
Minutia. (Valley is also referred as Furrow, Termination is

12
also called Ending, and Bifurcation is also called
Branch)
Figure 2.2.2
14
Verification vs. Identification
Figure 2.2.3
Simplified Fingerprint Recognition System

15
Figure 2.2.4
16
Minutia Extractor
Figure2.2.5
Figure 3.2.0
blueprint
Figure 3.2.1
Minutia Matcher
18
Data Model as a communication tool and database

21
Summary Of Steps In Exporting Data From Matlab To A
35
Database
Figure 4.0
Data Modeling Stage

42
Figure 4.1
Matlab Database Toolbox Stage
44
ABSTRACT
Fingerprint verification is one of the most reliable personal identification
methods in biometrics. This report presents the design and development of a
database for minutiae storage in a MATLAB-based fingerprint recognition
system. An examination authentication system (EAS) that will screen
students of Faculty of engineering at the entrance of an examination will also
be used as a case study for the database generated. The database will be
implemented by making use of MySQL, a relational database management
system. The process starts from connecting MATLAB to MySQL using the
Database ToolBox, Java Database Connectivity (JDBC) Driver and proceeds to
perform queries on and from MySQL using MATLAB.
CHAPTER ONE
INTRODUCTION
1.1 INTRODUCTION
This chapter will focus on background of this project, problem of the
manual system, the objectives, scope and significance of the project.
1.2 BACKGROUND
A fingerprint database image may readily be used as a background
database or may be used for testing purpose. Collecting large
databases of fingerprint images is problematic due to the great
amount of time and money required and the private legislation, which
in some countries prohibits the diffusion of such personal information.
Fingerprint recognition is one of the most important biometric
techniques. The largely accepted uniqueness of fingerprints and the
availability of low-cost acquisition devices make it well suited for a
wide range of application [1].
Great efforts are continually spent on designing new fingerprint
algorithms, both in academic and industrial environments, but not
enough care is usually dedicated to performance evaluation. In fact
testing an algorithm requires a large database of samples(thousands or
tens of thousands) due to small errors that have to be estimated.

Unfortunately, collecting large databases of fingerprint images is:
(1)expensive in terms of both money and time
(2)boring for both the people involved and the volunteers
(3)problematic due to the privacy legislation that protects such
personal data.
FVC 2000[2] and FVC 2002[2] are examples of technology evaluations;
where real fingerprint databases were collected to test and compare
different algorithm.
1.3 PROBLEM STATEMENT

Traditionally, the major technique for admitting students to an
examination over the years has been through the presentation of a
token e.g. ID cards, Dockets, etc whereby the student possesses a
physical and portable device which contains his/her identity. Students
authentication is done manually by using these tokens given by the
course adviser and not a system. This project presents a solution to the
problem of examination malpractices in universities or educational
system at large. With this manual system, there are some cases that
student can cheat by employing mercenaries that bear close
resemblance to them or by even swapping the passport photographs
on dockets or even by using fake clone of the stamp used on these
dockets. The invigilator cant monitor all the students in the
examination room. Because of this problem, a system may be needed
in order to enroll every students fingerprint at the point of admission
7
into the university. Matching will be done during the examination

period whereby each students fingerprint is taken at the point of entry
into the examination and is compared against the template already
stored in the students database.
1.4 OBJECTIVES
The objective is important to achieve the goal. The main objectives of
this project are:
1.4.1 To realize a database for minutiae storage using MATLAB and

MySQL taking examination authentication as a case study.
The first objective is to develop the database. This database will make
use of MySQL ,a relational database Management System (RDBMS)
which will be
integrated with a MATLAB algorithm for fingerprint
recognition from an earlier project [3]. This database will store minutiae
details extracted from a fingerprint image from another standard
database FVC2002
[2]
in the absence of a real students fingerprint
from a scanner/sensor. The database will also provide access to other

details to enable access to the information of the student easily.
1.4.2 To continue work on an earlier project
[3]
if possible.
Among the other future work recommended is to obtain a program that

will carry out the Alignment Stage of the minutia matching process and
automate the process completely using GUI and only enable button
clicks to carry out the whole processing.
1.5 PROJECT SIGNIFICANCE

8
The significance must be considered in every proposed system to

ensure that the proposed system will provide more benefit to all
parties especially the faculty. For this project, the significance has been
identified. The significance can be categorized into 3 groups. There
are:
1.5.1
Faculty
This significance of this system is mainly focusing for the faculty of
engineering itself. This is because this project will enhance the
efficiency of the manual system. It also can reduce in terms of docket
printing and avoid human error. All activities are done by computer and
technology.
1.5.2 Course Advisers

With this system, course advisers can also have access to the
students database which will consist of the fingerprint details and also
the courses for which the student will be writing examinations.
1.5.3 Students
As a user, students can motivate themselves to be more disciplined,
focused and dedicated.
1.6 CONCLUSION
This chapter discusses several important sections that relates to this and
it shall serve as a guideline to the successful completion of the project. It
includes elaboration on the background, problem statement, objectives
and significance of the project.
CHAPTER TWO
LITERATURE REVIEW
A fingerprint is an impression of the friction ridges of all or any part of
the finger. A friction ridge is a raised portion of the epidermis on the palmar
(palm and fingers) or plantar (sole and toes) skin, consisting of one or more
connected ridge units of friction ridge skin. These ridges are sometimes
known as "dermal ridges" or "dermal papillae.
Fingerprints may be
deposited in natural secretions from the eccrine glands present in friction

ridge skin (secretions consisting primarily of water) or they may be made by
ink or other contaminants transferred from the peaks of friction skin ridges to
a relatively smooth surface such as a fingerprint card. The term fingerprint
normally refers to impressions transferred from the pad on the last joint of
fingers and thumbs, though fingerprint cards also typically record portions of
lower joint areas of the fingers (which are also used to make identifications)
[4].
10
Figure2.0 A fingerprint image acquired by an Optical Sensor
2.1 BRIEF HISTORY OF FINGERPRINT

The basic idea of fingerprint started in the ancient Babylonian times when
Fingerprints were used on clay tablets for business transactions and also on
clay seals in ancient China.
The following gives some insight into the evolution of fingerprint into that
which we know it to be today:
In 14th century Persia, a doctor, observed on various official
government papers that no two fingerprints were exactly alike.

In 1686, Marcello Malpighi a professor of anatomy at the University of
Bologna, noted in his treatise; ridges, spirals and loops in fingerprints.
He made no mention of their value as a tool for individual
identification.
In 1880 (First Latent Print Identification) , Dr. Henry Faulds published
an article in the Scientific Journal, "Nature". He discussed fingerprints
as a means of personal identification, and the use of printers ink as a
11
method for obtaining such fingerprints. He is also credited with the

first fingerprint identification of a greasy fingerprint left on an alcohol
bottle.
Juan Vucetich made the first criminal fingerprint identification in 1892.
Sir Francis Galton published his book, "Fingerprints", establishing the
individuality and permanence of fingerprints. Galton's primary interest
in fingerprints was as an aid in determining heredity and racial
background. He was able to scientifically prove that fingerprints do not
change over the course of an individual's lifetime, and that no two
fingerprints are exactly the same. According to his calculations, the
odds of two individual fingerprints being the same were 1 in 64 billion.
Galton identified the characteristics by which fingerprints can be
identified. A few of these same characteristics (minutia) are basically
still in use today, and are sometimes referred to as Galton Details.

By 1946, the FBI had processed 100 million fingerprint cards in
manually maintained files; and by 1971, 200 million cards. With the
introduction of automated fingerprint identification system (AFIS)
technology, the files were split into computerized criminal files and
manually maintained civil files. Many of the manual files were
duplicates though, the records actually represented somewhere in the
neighborhood of 25 to 30 million criminals, and an unknown number of
individuals in the civil files.

Today, the largest AFIS repository in America is operated by the
Department of Homeland Security's US Visit Program, containing over
12
100 million persons' fingerprints, many in the form of two-finger

records. The two-finger records are non-compliant with FBI and
Interpol standards, but sufficient for positive identification and
valuable for forensics because index fingers and thumbs are the most
commonly identified crime scene fingerprints. The US Visit Program
has been migrating from two flat (not rolled) fingerprints to ten flat
fingerprints since 2007. "Fast capture" research funded by the US
government will enable implementation of ten "rolled print equivalent"
fingerprint recording (within 15 seconds per person fingerprinted) in
future years. The largest ten print AFIS repository in America is the
FBI's Integrated AFIS (IAFIS) in Clarksburg, WV. IAFIS has more than 60
million individual computerized fingerprint records (both criminal and
civil applicant records). Old paper fingerprint cards for the civil files
are still manually maintained in a warehouse facility (rented shopping
center space) in Fairmont, WV, though most enlisted military service
member fingerprint cards received after 1990, and all military-related
fingerprint cards received after 19 May 2000, have now been
computerized and can be searched internally by the FBI. In "Next
Generation Identification," the FBI may make civil file AFIS searches
available to US law enforcement agencies through remote interface.
All US states and many large cities have their own AFIS databases,
each with a subset of fingerprint records that is not stored in any other
database. Many also store and search palm prints. Law enforcement
13
fingerprint interface standards are important to enable sharing records

and reciprocal searches to identify criminals. [5]
2.2 FINGERPRINT RECOGNITION SYSTEM

Every person has minute raised ridges of skin on the inside surfaces of
his/her fingers, which display a number of characteristics known as minutiae
[6].
The minutiae do not change naturally during a persons life.
Figure 2.2.1. Minutia. (Valley is also referred as Furrow, Termination is

also called Ending, and Bifurcation is also called Branch)
These minutiae, the loops, whorls, arches, ridges, valleys, ridge endings,
etc., can be matched in two basic ways:
Identification (1: n, 1: many matching) of an individual using their
fingerprints is usually done for forensic or background checking purposes.
AFIS (Automated Fingerprint Identification System) is the most mature of all
biometric systems with the most widespread use, and has gone through its
own major advancements in terms of reliability and integration over the past
decade. This technology has traditionally been used by law enforcement
agencies around the globe to identify individuals. However, the technology is
increasingly being used in the private sector as a tool for employee
14
background checking, and its use is expanding within the public sector as
part of civil identification systems. AFIS systems require the collection of all
ten finger and thumb prints, both flat and rolled.
Key Applications:
Law Enforcement Forensics
Civil Identification
Background Checking
Employment
Adoption/Foster Parenting
Border Control/Visa Issuance
15
[7]
Figure 2.2.2 Verification vs. Identification

Verification (1:1 matching) is accomplished using systems which are
generally referred to as fingerprint Recognition systems. This usually
involves an individual pressing one or two fingers (or finger and thumb)
against a reader to verify that they are who they say they are (usually in
combination with a smart card which includes a stored template for that
individual). The technologies underlying fingerprint readers vary, including
optical, silicon, and ultrasound-based models. Fingerprint technologies have
been reduced in size that readers can now be incorporated into very small
form factors, embedded into laptops, mobile telephones or PDAs, even
integrated with portable USB drives, where they replace passwords as the
key to gain access. This technology is widely used for access control (both
physical and logical) applications. Fingerprint-based systems are commonly
16
combined with smart card technologies and boast high levels of accuracy.
Fingerprint recognition technologies do have some key drawbacks, however:
there continues to be limited interoperability, as well as some resistance
from users who may associate fingerprint technologies with law-enforcement
applications. With a large number of vendors competing for a piece of the
fast growth market for biometrics, the costs for use of fingerprint recognition
has been driven down to the point that it has become quite commonplace,
included as a standard option on a number of portable devices from
manufacturers. High profile government sponsored programs (e.g., US VISIT)
will likely force greater global adoption of fingerprint recognition technology
in passport and immigration programs.[7]
2.2.1 SYSTEM LEVEL DESIGN

A fingerprint recognition system constitutes of fingerprint acquiring device,
minutia extractor and minutia matcher [Figure 2.2.3].
Figure 2.2.3 Simplified Fingerprint Recognition System

17
For fingerprint acquisition, optical or semi-conduct sensors are widely used.

They have high efficiency and acceptable accuracy except for some cases
that the users finger is too dirty or dry. However, the testing fingerprint for
my project is from the available fingerprints provided by FVC2002
(Fingerprint Verification Competition 2002)[12] so no acquisition stage is
implemented. The minutia extractor and minutia matcher modules are
explained in detail in the next part for algorithm design.
2.2.2 ALGORITHM LEVEL DESIGN

To implement a minutia extractor, a three-stage approach is widely used by
researchers. They are preprocessing, minutia extraction and post processing
stage [Figure 2.2.4].
Figure 2.2.4 Minutia Extractor

The fingerprint image preprocessing stage
18
[3]
includes:
Image Enhancement, a necessary stage for fingerprint recognition

using minutiae extraction method which involves:
Block-wise Image Normalization: is a pixel-wise enhancement
preprocessing stage necessary to remove some noise due to
imperfections present in fingerprint reader/scanner after acquisition
of fingerprint image.
Orientation Estimation: The orientation field, which is a gradient
filed, defines the local orientation of the ridges contained in the
fingerprint, The orientation estimation is a required step in the
enhancement process by Gabor filtering which as a frequency
selective process also requires the ridge frequency.

Ridge frequency determination, and
Gabor filtering
Binarization: transform the 8-bit gray scale fingerprint image into a 1-
bit image with 0 values for ridges and 1 value for valleys, and
Segmentation: to separate the fingerprint which is the foreground from
the image background.
The minutiae extraction stage includes:
Thinning: Ridge thinning is used to eliminate the redundant pixels of
ridges till the ridges are just one pixel wide.

Minutiae marking: After thinning, then the minutiae points are marked.
For the post processing stage, a more rigorous algorithm is developed to

remove false minutia.[13] Also some of the earlier methods introduce some
spurious minutia points in the image and so to keep the recognition system
consistent these false minutiae need to be removed.
19
Figure2.2.5 Minutia Matcher

The minutia matcher chooses any two minutia as a reference minutia pair
and then matches their associated ridges first. If the ridges match well, two
fingerprint images are aligned and matching is conducted for all remaining
minutia [Figure 2.2.5].
2.3 MINUTIAE STORAGE

As seen from figure 2.2.2, there is a need to store minutiae in a database
from where it will now be used for matching and that is the real essence of
this project.
2.4 FINGERPRINT CLASSIFICATION

This is the process of classifying a given fingerprint into a number of classes
such as left loop, right loop, whorl, arch and tented arch, double loop, etc. It
20
should be noted that this classification of fingerprints is essentially

skewed[14]. A fingerprint classification system should be invariant to rotation,
translation,
and
elastic
distortion
of
the
frictional
skin.
classification is used for indexing in the fingerprint database.
Fingerprint
A typical
fingerprint classification approach is based on the extraction of fingerprint

singular points (core and delta) and the implementation of a rule based
classification system that takes its decision based on the number and
placement of core and delta in a fingerprint. Different classification
algorithms are proposed[15,16,17].
21
CHAPTER THREE
PROJECT IMPLEMENTATION
The project will be implemented using MySQL Server 5.1.33, MySQL
WorkBench 5.2.31 and also MATLAB 7.5.0 produced by Mathworks Company
Inc.
3.1 DESIGN STATEMENT

1. Read a fingerprint image from a file
2. Apply the algorithm to extract the minutiae points and remove
false/spurious minutiae points.
3. Store the minutiae points as a template to be saved in a MySQL database.
4. Repeat steps 1-3 for as many fingerprint images. Thus, a database is
achieved/created.
5. Read a new fingerprint image.
6. Perform Step 2 on the image.
7. Then compare with the database.
SOLUTION METHOD
22
Here, minutiae points shall be saved in a MySQL database rather than in a

text file as in the past algorithm
[3]
. The approach in doing this should
involve:
1. Connecting Matlab Database ToolBox to MySQL
2. Data Modeling: conceptual representation of the data structures that
are required by the database.
3. Editing the algorithm to save minutiae in MySQL database.
4. Then apply the steps in the design statement above.
3.2 DATA MODELING

A data model is a device that helps the users or stakeholders understand
clearly the database system that is being implemented based on the
information requirements of an organization, and enables the database
practitioners to implement the database system exactly conforming to
the information requirements. A data model, therefore, serves as a
critical tool for communication with the users; it also serves as a blueprint
of the database system for the developers.
23
Figure 3.2.0 Data Model as a communication tool and database blueprint.
3.2.1
FACTORS THAT ENSURE A GOOD DATA MODEL
-Completeness: A good model should support all the necessary data.

-Non-redundancy: A good model should specify a database in which the
same fact could be recorded more than once.
-Enforcement of Business Rules: A good model must accurately reflect
and enforce the rules that apply to the business data.
-Data Reusability: In a good model, the data stored in the database should
be re-useable for purposes beyond those anticipated in the process model.
Once an organization has captured data to serve a particular requirement,
other potential uses and users almost invariably emerge.
-Stability and Flexibility: A good model should cope well with possible
changes to the business requirements and new data required to support such
24
changes should be accommodated in existing tables. Alternatively, simple

extensions will suffice or major structural changes will have to be made with
corresponding impact on the rest of the system. All these largely determine
how quickly the system can respond to business change, which, in many
cases, determines how quickly the business as a whole can respond. A data
model is stable in the face of a change to requirements if we do not need to
modify it at all. A data model is flexible if it can be readily extended to
accommodate likely new requirements with only minimal impact on the
existing structure.
-Elegance: A good data model should provide a reasonably neat and simple
classification of the data as the difference in development cost between
systems based on simple, elegant data models and those based on highly
complex ones can be considerable indeed.
-Communication: A good data model should be effective in supporting
communication among the various stakeholders in the design of a system
(i.e. the tables and columns represent business concepts that the users and
business specialists are familiar with and can easily verify) and as such the
programmers will interpret the model correctly. The quality of the final model
will depend very much on informed feedback from business people.
Programmers, in turn, need to understand the model if they are to use it as
intended.
-Integration: The proposed database should fit with the organizations
existing and future databases. Even when individual databases are well
25
designed, it is common for the same data to appear in more than one
database and for problems to arise in drawing together data from multiple
databases.
-Conflicting Objectives: In many cases, the above factors will conflict with
one another. An elegant but radical solution may be difficult to communicate
to conservative users. We may be so attracted to an elegant model that we
exclude requirements that do not fit. A model that accurately enforces a
large number of business rules will be unstable if some of those rules
change. And a model that is easy to understand because it reflects the
perspectives of the immediate system users may not support reusability or
integrate well with other databases.
Our overall aim is to develop a model that
provides the best balance among these possibly conflicting objectives. As in
other design disciplines, achieving this is a process of proposal and
evaluation, rather than a step by-step progression to the ideal solution.
3.2.2 LEVELS OF DATA MODELING

External Data Model: The external data model is a depiction of the
database system
from the viewpoints of individual user groups. This model may be used for
communication with individual groups of users. Each individual user group is
interested in a set of data items for performing its specific business
functions. The set of data items relevant for a specific user group forms part
of the external data model for this particular user group.
26
Conceptual Data Model: The conceptual data model is at a high and

general level,
intended mainly as a communication tool with the user community. In the
model, there is no room for details of data structure or for any considerations
of hardware and database software. This model does not even address
whether the final database system is going to be implemented as a relational
database system or any other type of database system. However, the model
should be complete and include sufficient components so that it would be a
true representation of the information requirements of the organization. . It is
the focus of communication between the data modeler and business
stakeholders, and it is usually presented as a diagram with supporting
documentation.
Logical Data Model: In a sense, the logical data model for an organization
is the aggregation of all the parts of the external data model. In the above
external data model, three user groups are shown. We assume that there are
only three user groups in the organization. Therefore, the complete logical
model must represent all the combined information requirements of these
three user groups. For the relational type of database system, the logical
model represents the information requirements in the form of twodimensional tables with rows and columns. The logical data model is a
translation of the conceptual model into structures that can be implemented
using a database management system (DBMS). Today, that usually
means that this model specifies tables and columns, as we saw in our first
27
example. These are the basic building blocks of relational databases, which
are implemented using a relational database management system
(RDBMS).
Physical Data Model: A physical data model has little use as a means of
communication with the users. Its primary purpose is to act as a blueprint for
the implementation of the database system. The details contained in a
physical data model are beyond the normal comprehension of the users. The
model expresses too many intricate details. It includes considerations of the
particular DBMS and the hardware environment in which the database
system gets implemented. The physical data model incorporates any
changes necessary to achieve adequate performance and is also presented
in terms of
tables and columns, together with a specification of physical
storage (which may include data distribution) and access mechanisms.

Different methodologies differ on the exact level of detail that should
be included in each model and at what point certain decisions should be
taken. In some methodologies, the translation from conceptual to logical is
completely mechanical; in others, including our recommended approach,
there are some decisions to be made. The step from logical to physical may
be straightforward with no changes to tables and columns, if performance is
not a problem, or it may be highly complex and time-consuming, if it
becomes necessary to trade performance against other data model quality
criteria.
3.2.3 ADVANTAGES OF USING THE RELATIONAL MODEL

28
It describes data independent of the actual physical representation of the

data.
The model of data is simple and easy to understand.
It provides high-level operations for querying the data.
The model is conceptually simple, allowing application programmers to be
able to quickly grasp the important concepts they need to get started with
their work.
The model is based on a mathematical structure, which allows many
operational aspects to be proved, and the operations have well-defined
properties.
Its easy to modify and add to relational databases.
The same database can be represented with less redundancy.
3.3 NORMALIZATION
This is a set of rules for allocating data to tables in such a way as to
eliminate certain types of redundancy and incompleteness.
[20]
It is also a
data modeling technique, the goal of which is to organize data elements in

such a way that theyre stored in one place and one place only (with the
exception of foreign keys, which are shared). In practice, normalization is
usually one of the later activities in a data modeling project, as we cannot
start normalizing until we have established what columns (data items) are
required. In another approach, normalization is used in the logical database
design stage, following requirements analysis and conceptual modeling.
29
A normal form is a state of a relation that can be determined by

applying simple rules regarding dependencies to that relation. Normal forms
are designed to prevent update and delete anomalies, data redundancies,
and inconsistencies.
[19]
Universal properties
No duplicate members of the set.

Record order unimportant (top to bottom).
Attributes order unimportant (left to right).
All attribute values are atomic.
No single attribute is allowed to hold more than
one value at one time.
First Normal Form (1NF)
The appropriateness of the primary key.
Second
Normal
Form The dependence of all attributes on all aspects of
(2NF)
the primary key.
Third Normal Form (3NF)
The dependence of any attribute on any attribute
other than the primary key.
Boyce-Codd Normal Form

(BCNF)
Fourth
(4NF)
Normal
Verifies that all data sets are identified and

segregated.
Form Verifies that all attributes are single valued for a

member of the set.
Domain Key Normal Form Verifies that all constraints are the logical
(DKNF)
consequence of the definition of the keys and the
domains (data value rules).
Table 1.1: Normal Forms

Universal Properties of Relations
30
The universal properties of relations are the preconditions that must be in

place prior to the test for normal forms. They refer to that two-dimensional
math form called a relation upon which relational design is based.
There must be no duplicate instances (duplicate members of the set).
Instances are unordered (top to bottom).
Data elements are unordered (left to right).
All data element values are atomic. No single column describing a single
instance is allowed to hold multiple values at one time.
STEPS OF NORMALIZATION
STEP 1: Put the data into tabular form (by removing repeating groups) and
remove duplicated data to separate tables.
COURSE CODE
COURSE
DEPARTMENT
EEG 502
TITLE/DECSRIPTION
DIGITAL COMPUTER
ELECT./ELECT.
DESIGN
EEG 520
DIGITAL CONTROL
Table 1.2: Course Table
ELECT./ELECT.
This table has 3 columns/fields and the course code can be set as primary
key as it contains different courses.
NAME
AKANGBE SAIDAT
OMOTOLA
BELLO
MATRIC. NO.
DEPARTMENT
DATE OF
070408006
COMPUTER
BIRTH
26/06/1990
060408029
ENGINEERING
ELECT. /ELECT.
11/03/1987
NASIRULLAHI
ENGINEERING
31
BOLARINWA
Table 1.3: Student Table
This table contains 4 columns that stores information about each
student and here we have set the Matric No as Primary key because the
table contains information about different students. Here, the Matric No
serves as a unique identifier for each student.
NAME
AKANGBE SAIDAT
MATRIC. NO.
070408006
DEPARTMENT
COMPUTER
COURSE
EEG 502
OMOTOLA
BELLO
060403029
ENGINEERING
ELECT. /ELECT.
EEG 502
NASIRULLAHI
ENGINEERING
BOLARINWA
Table 1.4: Registration/Exam Enrollment Table
This details contains the registration details of students writing
examinations for different courses. A student can offer many courses,
therefore matric no may not be a primary key. Also students from many
department can/might offer a course.
MATRIC
NO.
NO.
TERMS
ANGL
BIFS
ANGL
ANGL
ANGL
NO
OF
OF
BIF
ORIENTATION
X
Y
TERM
ORIENTATION
X
Y
S
2
S
2
0704080
27
27
-1.85
172
-2.07
139
-1.67
24
184
NaN
NaN
0.00
191
-2.62
NaN
92
-2.36
1.57
06
27
0604030
66
32
33
67
-0.52
-1.05
29
69
156
1.02
67
177
-2.36
2.09
0.52
75
120
-1.68
74
135
2.09
-1.05
0.00
Table 1.5: Minutiae/Finger Details Table

From the algorithm after minutiae orientation, the minutiae details to
be saved in our database include:
-Number of Terminations
- Number of Bifurcations
-Termination Orientation
-Termination Angle of Orientation
-Bifurcation orientation
- Bifurcation Angle of Orientation
The table needs to be normalized as it has not yet satisfied the
universal properties of a relation/table.
STEP 11: Considering the Minutiae table, I determined the columns by:
(i) Putting one fact per column
-Looking at the terms orientation column in table 1.5 ,we can see that it
contains two distinct facts i.e X and Y and the same holds for the bifs
orientation table. Rather, these two facts should be recorded in separate
columns.
(ii) Hidden Data
-The second piece of tidying up involves making sure that we have not lost
any data in the translation to tabular form. The most common problem here
is that we cannot rely on the rows of the table being stored in any particular
33
order. After scrutinizing the nature of our database, the hidden data to
include is FINGERTYPE. This is because the design allocate for different kinds
of finger e.g thumb, index Finger, Middle Finger, Ring Finger & Little/Pinky
finger for either left or right hand.
(iii) Determining the Primary Key
- Finally, we determine a primary key for the table. The choice of primary
keys is a critical (and sometimes complex) task. For the moment, we will
simply note that the primary key is a minimal set of columns that contains a
different combination of values for each row of the table. Another way of
looking at primary keys is that each value of the primary key uniquely
identifies one row of the table. In this case, Course Code and Matric No will
do the job as stated earlier.
Primary Keys
A primary key is a nominated column or combination of columns that
has a different value for every row in the table.
Candidate Keys
Sometimes more than one column or combination of columns could
serve as a primary key.
Foreign Keys
When we remove repeating groups to a new table, we carry the
primary key of the original table with us, to cross-reference or point back to
the source. In moving from first to third normal form, we left determinants
behind as cross-references to the relevant rows in the new tables. These
34
cross-referencing columns are called foreign keys, and they are the principal
means of linking data from different tables.
3.4USING MATLAB DATABASE TOOLBOX

The Database Toolbox product enables you to exchange data between
MATLAB and a database . With the Visual Query Builder graphical user
interface tool within the toolbox, you can select data from database, use
MATLAB tools to analyze data, and store results back in the database.
However, the Database Toolbox software is not suitable for database
administration. When performing administrative things (e.g. creating tables),
use database management system application (e.g. mysql client program)
After opening the MATLAB workspace window , Open the database toolbox
graphical user interface from the lower left corner of workspace window:
Start|Toolboxes|Database|Visual Query Builder (querybuilder)
[11]
The
first thing to do is to define data source (the data itself and the connection to
the data): Do the following in the Visual Query Builder window
1. Click Query | Define JDBC Data Source
2. Click Create New File in the opened window
3. Give a name to the data source MAT file and click Save. This file saves
your data source specification, which you can use in the future sessions
35
4.After that enter information for your data source into the Name, Driver, and
URL fields as follows:
o Name: database_name_you_want_to_connect
o Driver: com.mysql.jdbc.Driver
o URL:jdbc:mysql://localhost:3306/database_name_you_want_to_co
nnect
5. Click Add/Update and OK
Now you can start using Visual Query Builder to import or export data
between MATLAB and database. In order to import data from a database into
the MATLAB workspace, do the following in the Visual Query Builder window:
1. Click the Select button under Data operation
2. Select your database from the data source list. Give your username and
password
3. Select tables and fields which you want to query
4. Enter a suitable name (e.g. A) in the MATLAB workspace variable field.
This variable stores the data that the query returns
5. Click Execute to run the query. The query result appers in the Data field.
6. Double-click the variable in the Data field. That opens a Variable Editor,
where the query data is shown.
In order to export data from the MATLAB workspace into a database, do the
following in the Visual Query Builder window:
36
1. Click the Insert button under Data operation

2. Select your database from the data source list
3. Select tables and fields which you want to export data from the MATLAB
software
4. Do the following in the MATLAB workspace (in the command window), not
in the Visual Query Builder window: Attach the data to a variable:
variable_name = \{'data_value'\}; and press enter.
5. Enter the same name of the variable in the MATLAB workspace variable
field (in the Visual Query Builder window)
6. Click Execute to export the data.
Alternatively, you can use the coding approach of connecting MATLAB to a
MySQL database. This is preferable as you could just add this code to your
algorithm or project.
% Database Server
host = 'localhost:3306';
% Database Username/Password
user = 'root';
password = 'password';
% Database Name
dbName = 'project';
%Enter the username of your account

%Enter the password to your account here.
%Enter the name of your database e.g project
% JDBC Parameters
jdbcString = sprintf('jdbc:mysql://%s/%s', host, dbName);
jdbcDriver = 'com.mysql.jdbc.Driver';
37
% Set this to the path to your MySQL Connector/J JAR

%javaaddpath('C:\mysql-connector-java-5.1.14-bin.jar')
% Create the database connection object
dbConn = database(dbName, user, password, jdbcDriver, jdbcString)
isconnection(dbConn)
QUERY = 'INSERT INTO student(deptID) values(04)';
%type your query here
[filenames]=fetch(dbConn,QUERY)
close(dbConn);
Figure 3.2.1: Summary Of Steps In Exporting Data From Matlab To A

Database
All these was carried out to successfully connect MATLAB to MySQL.
38
CHAPTER FOUR
RESULTS AND DISCUSSIONS
After each stage in the solution method was carried out, different results
were obtained. The results achieved are shown for the different stages.
NORMALIZATION STAGE
- After normalizing, the following tables were obtained:
Course
Course Title
Code
Semest
Department
Unit
er
Table 1.6: Course Table (Final)

It was then created in the MySQL database by performing the following
queries:
CREATE TABLE `course` (
`Course Code` varchar(6) NOT NULL,
`Course Title` varchar(45) NOT NULL,
`Semester` varchar(1) NOT NULL,
`Department` varchar(45) NOT NULL,
Ùnit` smallint(6) NOT NULL,
PRIMARY KEY (`Course Code`,`Course Title`,`Semester`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1
39
Here, Course Code, Course Title and semester are all primary
keys/candidate keys because the design assumes that there can be only one
unique course code with a particular title that can be offered in a semester.
Matric
Number
No
of
Angle
Angle
Angle
Finger
Type
Bifurcatio
ns
Table 1.7: Bifurcations Table (Final)
It was also created in the MySQL database by performing the following
queries:
CREATE TABLE `bifurcations` (
`X` int(3) NOT NULL,
`Y` int(3) NOT NULL,
Àngle1` decimal(3,2) NOT NULL,
`Finger Type` varchar(20) NOT NULL,
`Matric No` varchar(9) NOT NULL,
`Number of Bifurcations` int(11) NOT NULL,
PRIMARY KEY (`Matric No`,`X`,`Y`,Àngle1`,Àngle2`,Àngle3`,`Finger
Type`,`Number of Bifurcations`)
KEY `Matric No` (`Matric No`),
KEY `Finger Type` (`Finger Type `),
40
CONSTRAINT ` Finger Type ` FOREIGN KEY (`Finger Type `) REFERENCES `finger`

(`Finger Type `) ON DELETE CASCADE ON UPDATE CASCADE,
CONSTRAINT `Matric No` FOREIGN KEY (`Matric No`) REFERENCES `student`
(`Matric No`) ON DELETE CASCADE ON UPDATE CASCADE
Here, we have made the students Matric No, bifurcation positions(X,

Y), bifurcation orientation (Angle1, Angle2, Angle3), Finger Type and number
of bifurcations candidate keys because our design assumes that the
probability of 2 persons out of about 6 billion people having the same
bifurcation position and orientation is very small and hence negligible for a
small population i.e. Students as used in this project.
Matric
Number of
No
Terminatio
Angle
Finger
Type
ns
Table 1.8: Terminations Table (Final)
The MySQL queries to create this is:
CREATE TABLE `terminations` (
Àngle` decimal(3,2) NOT NULL,
`Number of Terminations` int(11) NOT NULL,
PRIMARY KEY (`Matric No`,`X`,`Y`,Àngle`,`Finger Type`,`Number of
Terminations`)
41

Here, the case of candidate keys is exactly the same as that of

bifurcations and it has been assumed in the design that that the probability
of 2 persons out of about 6 billion people having the same bifurcation
position and orientation is very small and hence negligible for a small
population i.e. Students as used in this project.
Matric
First
Last
Other
Date of
Departme
Phone
No
Name
Name
Names
Birth
nt
No
Table 1.9: Students Table (Final)

The MySQL queries to create this include:
CREATE TABLE `student` (
`First Name` varchar(25) NOT NULL,
`Last Name` varchar(25) NOT NULL,
Òther Names` varchar(25) NOT NULL,
`Date of Birth` date NOT NULL,
`Phone No` varchar(18) DEFAULT NULL,
`Sex` char(1) NOT NULL,
PRIMARY KEY (`Matric No`)
42
Sex
Finger Type
Table 2.0: Finger Table (Final)
CREATE TABLE `finger` (
PRIMARY KEY (`Finger Type`)
Matric
Course Code
Semester
Session
No
Table 2.1: Registration Table (Final)
CREATE TABLE `registration` (
`Session` year(4) NOT NULL,
PRIMARY KEY (`Matric No`,`Course Code`,`Session`),
KEY `Course Code` (`Course Code`),
CONSTRAINT `Course Code` FOREIGN KEY (`Course Code`) REFERENCES `course`
(`Course Code`) ON DELETE CASCADE ON UPDATE CASCADE,
43
As seen from Table 1.7 and 1.8, the minutiae table in table 1.5 has
been split into two tables i.e. Bifurcations and Terminations. Also, extra
details were added in the Students Table e.g.
-First Name, last Name, other names: used to distinguish or almost uniquely
identify candidates for the exam as against name only as used in the
former table. Also, name is a reserved word for most databases so this
helps us to classify. ,
-Phone number, Sex, etc.
DATA MODELLING STAGE
The relational model obtained is as shown below:
44
Figure 4.0: Data Modeling Stage

This was obtained using the MySQL WorkBench, after creating the
schemas i.e. tables.
: This icon stands for a 1: n relationship.

-Matric No under Student Table has a 1: n relationship with Matric No under
Terminations table. The same applies for Matric No under Bifurcations table.
45
- Finger Type under finger Table has a 1: n relationship with Finger Type under
Terminations table. The same applies for Finger Type under Bifurcations
table.
- Course Code under Course Table has a 1: n relationship with Course Code
under Registration table.
N.B: Relationships illustrate how two entities share information in the
database structure. Entities with one-to-one relationships should be merged
into a single entity.Each remaining entity is modeled by a table with a
primary key and attributes, some of which may be foreign keys. One-to-many
relationships are modeled by a foreign key attribute in the table representing
entity on the "many" side of the relationship. Many-to-many relationships
among two entities are modeled by a third table that has foreign keys that
refer to the entities. These foreign keys should be included in the relationship
table's primary key, if appropriate.
MATLAB DATABASE TOOLBOX STAGE

After successfully connecting to the Matlab Database ToolBox
otherwise known as Query Builder, we have:
46
Figure 4.1: Matlab Database Toolbox Stage

Then I went ahead to perform queries to export data embedded in our
MATLAB variables to our MySQL database named project. The variables
were obtained after running the MATLAB codes in matching algorithm by the
initial author.
[3]
47
CHAPTER FIVE
CONCLUSION
This project report has briefly introduced the major concepts in
database designing for storing minutiae in a fingerprint recognition system
and has succeeded in overcoming the challenges faced by the initial author.
A significant improvement was made on the data modeling stage with the
introduction of normalization approach which is the most versatile technique
used in data modeling.
In addition, minutiae details i.e. terminations and bifurcations were
successfully separated into different schemas/tables. In order to ensure
connection between the matching algorithm and the DBMS being used, the
driver meant for that purpose had to be carefully configured.
Queries were then being made to the database to create the schemas
obtained after modeling and also to export data from the algorithm to the
database for storage purposes and for future reference.
A graphical user interface was also obtained to view the processing stages
with the use of the GUI ToolBox in MATLAB (See AppendixB), thereby
automating the process with the use of enabled buttons clicks only.
5.0 CHALLENGES
Several challenges were encountered while carrying out this project and
these include:
48
Difficulty in setting up connection between MATLAB and MySQL using

JDBC driver: I spent a very long time in doing this because it required
case-sensitive strings that had to be used in the configuration such
that even a mere mistype of a character e.g. d instead of D
prevented a successful connection.

Difficulty in determining how to model the minutiae details into a
database: this was because of the long nature of the orientations and
positions of the bifurcations and terminations as one finger could e.g.
have like 35 terminations and 53 bifurcations.
5.1
FUTURE WORK
Although, this project has tried its best in storing minutiae in a
database, there is still need for some future work to be carried out.
This include:
1) The possibility of encrypting all the minutiae details in a long string
as used in the SECUGEN fingerprint reader: this might involve some
form of coding/hashing.
49
REFERENCES
[1 ]. Nalini Ratha, Ruud Bolle(2004) , Automatic Fingerprint Recognition
System, New York: Springer-Verlag.
[2]. D. Maltoni, D. Maio, A.K. Jain & S. Prabhakar (2009), Handbook of
Fingerprint Recognition (Second Edition), London: Springer.
[3].Adeoye Adewale (2010), Matlab-based design and implementation of
fingerprint matching algorithm, University of Lagos, Nigeria.
[4]. Sy Mohd Syathir bin Sy Ali Zainol Abidin (2007) , Intelligent Fingerprint
Recognition System , Universiti Malaysia Pahang
[5]. Greg Moore , David L. von Minden, PhD (2011) , The History of
Fingerprints ,Retrieved March 8,2011, from
http://www.onin.com/fp/fphistory.htm
[6]. J. D.Woodward, N.M. Orlans, & P. T. Higgins (2002), Biometrics, New York:
McGraw-Hill.
[7]. Global Security Intelligence (2011), Fingerprint Recognition, Retrieved
from http://globalseci.com/?page_id=39
[8]. Wikipedia (2010), Database, http://wikipedia.org
50
[9]. J. Krupa, A. Prochazka, V. Hanta, R. Hava, Technical Computing Using

Sybase Database For Biomedical Signal Analysis, Institute of Chemical
Technology, Prague :
Department of Computing and Control Engineering
[10]. University Of Ibadan (2005) , MySQL Tutorial , ATDB.
[11].Math Works (2011),Database Toolbox
,http://www.mathworks.com/access/helpdesk/help/toolbox/database/gs/braie
y2-1.html#braiey2-24
[12] FVC2002, Finger Print Verification competition,
http://bias.csr.unibo.it/fvc2002/
[13] WU ZHILI (2002), Fingerprint Recognition, Hong Kong Baptist University.
[14]. Anil Jain, Sharath Pankanti (1998), Fingerprint Classification and
Matching,
Michigan State University, MI.
[15]. Andrew Senior (2001), IEEE Transactions on Pattern Analysis and
Machine
Intelligence, A Combination Fingerprint Classifier, vol. 23 ,no. 10, pp 11651174.
[16]. Meltem Ballan et al., A Fingerprint Classification Technique Using
Directional Images, Yildiz Technical University, Istanbul, Turkey and the
University of Texas at Austin , Austin, Texas.
51
[17]. A. Senior (1997), A Hidden Markov Model Fingerprint Classifier, Proc.

Asilomar
Conn. Signals, Systems, and Computers.
[18]. Paulraj Ponniah (2007), Data Modeling Fundamentals A practical Guide
for IT
Professionals, New Jersey: Wiley.
[19]. Sharon Allen & Evan Terry(2005),Beginning Relational Data
Modeling(Second
Edition), Apress.
[20]. Graeme C. Simsion & Graham C. Witt (2005), Data Modeling Essentials
(Third Edition), Morgan Kaufmann
APPENDIX A
BRIEF
DEFINITION
OF
SOME
KEY
COMPONENTS
AND
OPERATIONS
Database: A database consists of an organized collection of data for one
or more uses, typically in digital form.
[8]
One way of classifying databases
involves the type of their contents, for example: bibliographic, documenttext, statistical. Digital databases are managed using database management
systems, which store database contents, allowing data creation and
maintenance, and search and other access. Database architecture consists of
three levels, external, conceptual and internal. Biometric data storage, that
refers to the design and implementation of a relational database system. The
52
advantage of storing the biometric data in the database is based on the

inherent advantage presented by database systems: centralized accurate,
highly secure, reliable and organized data, easy access to insert, update and
retrieve biometric data.
DATABASE MANAGEMENT SYSTEM (DBMS): is a set of computer
programs that controls the creation, maintenance, and the use of a
database. It allows organizations to place control of database development in
the hands of database administrators (DBAs) and other specialists. A DBMS
is a system software package that helps the use of integrated collection of
data records and files known as databases. It allows different user
application programs to easily access the same database. DBMSs may use
any of a variety of database models, such as the network model or relational
model.
Open Database Connectivity (ODBC): this provides a standard

software interface for accessing a database management system (DBMS).
ODBC provides a universal middleware layer between the application and the
DBMS, allowing the application developers to only have to learn a single
interface, nor do they have to update their software if changes are made to
the DBMS specification, only the driver needs updating. An ODBC driver can
thus be thought of as analogous to a printer or other driver, providing a
standard set of calls for the application to use, which then translates those
commands into the correct commands at the DBMS end. An application that
can communicate through ODBC is referred to as ODBC-compliant. Any
53
ODBC-compliant can access any DBMS that has a corresponding driver. For
the driver, the ODBC model allows for two different solutions, either having
the driver reside on the client machine, or as a part of a server-side solution.
Drivers exist for all major DBMS.
Java Database Connectivity (JDBC): is an API (Application

Programming Interface) for the Java programming language that defines how
a client may access a database. It provides methods for querying and
updating data in a database. JDBC is oriented towards relational databases. A
JDBC-to-ODBC bridge enables connection to any ODBC-accessible data
source in the Java Virtual Machine (JVM) host environment. JDBC drivers are
client-side adapters (installed on the client machine, not on the server) that
converts requests from the Java programs to a protocol that the DBMS can
understand.
Metadata: Database store user data and they also store information
about the database itself. Most DBMSs have a set of system tables, which list
tables in the database, column names in each table, primary keys, foreign
keys, stored procedures, and so forth. Each DBMS has its own functions for
getting information about table layouts and database features. JDBC provides
the interface Database Metadata, which a driver writer must implement so
that its methods return information about the driver and/or DBMS for which
the driver is written. For example, a large number of methods return whether
or not the driver supports a particular functionality. This interface gives users
54
and tools a standardized way to get metadata. In general, developers writing

tools and drivers are the ones most likely to be concerned with metadata.
Matlab Database Toolbox: The Database Toolbox is one of an

extensive collection of toolboxes for use with MATLAB. The Database Toolbox
enables you to move data (both importing and exporting) between MATLAB
and popular relational databases.
With the Database Toolbox, you can bring data from an existing database
into MATLAB, use any of the MATLAB computational and analytic tools, and
store the results back in the database or in another database. Also, you can
read from the database, importing the data into the MATLAB workspace. You
can interact with most popular databases from within MATLAB, including
Oracle, Sybase, Microsoft SQL Server, and Informix. The toolbox also enables
simultaneous access to multiple databases within a single MATLAB session
and enables transactions involving large data sets. Using the Visual Query
Builder, you can interact with a database without knowing SQL. MATLAB
Database Toolbox
[10]
supports communication using ODBC or JDBC driver
with compatible database including IBM DB2, IBM Informix, Ingres, Microsoft
Access, Microsoft Excel, Microsoft SQL Server, MySQL, Oracle, PostgreSQL,
Sybase SQL Anywhere and Sybase SQL Server.
The
Math Works
Fig
3.2.1
MATLAB
Database
Schema
Toolbox
MySQL
JDBC
Driver
MATLAB-MySQL
55
communication
MySQL: is a relational database management system (RDBMS) that runs as

a server providing multi-user access to a number of databases. The database
(the "DB" in RDBMS) is the repository for the information you want to store,
structured in a simple, regular fashion:
o
o
o
o
The collection of data in a database is organized into tables.

Each table is organized into rows and columns.
Each row in a table is a record.
Records can contain several pieces of information; each column in a table
corresponds to one of those pieces.
The management system (the "MS") is the software that lets you use
your data by allowing you to insert, retrieve, modify, or delete records.
[10]
The word "relational" (the "R") indicates a particular kind of DBMS, one
that is very good at relating (that is, matching up) information stored in one
table to information stored in another by looking for elements common to
each of them. The power of a relational DBMS lies in its capability to pull data
from those tables conveniently and to join information from related tables to
produce answers to questions that can't be answered from individual tables
alone.
PRELIMINARY REQUIREMENTS FOR USING MYSQL ON YOUR MACHINE
You need to have the MySQL software installed. (e.g. with XAMPP server)
and configured
MySQL WorkBench(optional)
You need a MySQL account so that you can connect to the server.
You need a database to work with.
SQL QUERIES
56
CREATE TABLE `bifurcations` (

`Number of Bifurcations` int(11) NOT NULL,
PRIMARY KEY (`Matric No`,`X`,`Y`,Àngle1`,Àngle2`,Àngle3`,`Finger
Type`,`Number of Bifurcations`)
-----------------------------------------------------------------------------------------------------------CREATE TABLE `course` (
`Course Title` varchar(45) NOT NULL,
Ùnit` smallint(6) NOT NULL,
PRIMARY KEY (`Course Code`,`Course Title`,`Semester`)
----------------------------------------------------------------------------------------------------------CREATE TABLE `finger` (
57
PRIMARY KEY (`Finger Type`)

----------------------------------------------------------------------------------------------------------CREATE TABLE `registration` (
`Session` year(4) NOT NULL,
PRIMARY KEY (`Matric No`,`Course Code`,`Session`),
KEY `Course Code` (`Course Code`),
CONSTRAINT `Course Code` FOREIGN KEY (`Course Code`) REFERENCES `course`
(`Course Code`) ON DELETE CASCADE ON UPDATE CASCADE,
-----------------------------------------------------------------------------------------------------------CREATE TABLE `student` (
`First Name` varchar(25) NOT NULL,
`Last Name` varchar(25) NOT NULL,
Òther Names` varchar(25) NOT NULL,
`Date of Birth` date NOT NULL,
`Phone No` varchar(18) DEFAULT NULL,
`Sex` char(1) NOT NULL,
PRIMARY KEY (`Matric No`)
-----------------------------------------------------------------------------------------------------------58
CREATE TABLE `terminations` (

Àngle` decimal(3,2) NOT NULL,
`Number of Terminations` int(11) NOT NULL,
PRIMARY KEY (`Matric No`,`X`,`Y`,Àngle`,`Finger Type`,`Number of Terminations`)
APPENDIX B: GRAPHICAL USER INTERFACE

Among the future work recommended by the earlier author [3] is to
automate the process of the algorithm completely. Though some measure of
automation is observed especially as functions are used throughout in the
project, it was expected that future work will, perhaps, use a GUI and only
enable button clicks to carry out the whole processing.
Hence, I was able to obtain the following:
59
Figure B1: load FingerPrint image
60
Figure B2: Enhancement by Histogram Equalization
61
Figure B3:Enhancement by FFT
Figure B4: Adaptive Binarization after FFT

62
Figure B5: Orientation Flow Estimate
Figure B6: region of Interest (ROI)

63
Figure B7: Thinned-ridge map
Figure B8: remove H breaks

64
Figure B9: remove spike
65
Figure B10: Minutia
Figure B11: Extracting minutia
Figure B12: Saving Minutia in an ASCII(.dat) file

66
Figure B13: Opening Fingerprint Template files for matching
Figure B14: Matching Successful (100%)

The MATLAB code used for GUI include :
67
%gui_start.m
%Script file to generate the GUI
%Final Year Project 2010-2011
%Akangbe S.O
%Computer Engineering, UNILAG
clear
FigWin = figure('Position',[50 -50 650 500],...
'Name','Fingerprint Matching Algorithm - 070408006',...
'NumberTitle','off',...
'Color',[ 0.827450980392157 0.815686274509804 0.776470588235294 ]);
AxesHandle1 = axes('Position',[0.2 0.15 0.35 0.7],...
'Box','on');
AxesHandle2 = axes('Position',[0.6 0.15 0.35 0.7],...
'Box','on');
BackColor = get(gcf,'Color');
%[ 0.827450980392157 0.815686274509804 0.776470588235294 ]
%[ 0.741176470588235 0.725490196078431 0.658823529411765 ]
FrameBox = uicontrol(FigWin,...
'Units','normalized', ...
'Style','frame',...
'BackgroundColor',[ 0.741176470588235 0.725490196078431 0.658823529411765 ],...
'ForegroundColor',[ 0.741176470588235 0.725490196078431 0.658823529411765 ],...
'Position',[0 0 0.15 1]);
%create static text.
Text2 = uicontrol(FigWin,...
'Style','text',...
'Position',[0 0.95 1 0.05],...
'FontSize',15,...
'HorizontalAlignment','right', ...
'String','Fingerprint Matching Algorithm - 070408006');
Text2 = uicontrol(FigWin,...
'Style','text',...
'Position',[0 0 1 0.05],...
'FontSize',15,...
'HorizontalAlignment','right', ...
'String','Fingerprint Matching Algorithm - 070408006');
w=16;
textLoad='Load Fingerprint Image';
h=uicontrol(FigWin,...
'Style','pushbutton',...
68
'Position',[0,320,80,20],...
'String','Load',...
'Callback',...
['image1=loadimage;'...
'subplot(AxesHandle1);'...
'imagesc(image1);'...
'title(textLoad);'...
'colormap(gray);']);
text_filterArea='Orientation Flow Estimate';
'Position',[0,240,80,20],...
'String','Direction',...
'Callback',...
['subplot(AxesHandle2);[o1Bound,o1Area]=direction(image1,16);title(text_filterArea);']);
text_ROI='Region Of Interest(ROI)';
'Position',[0,220,80,20],...
'String','ROI Area',...
'Callback',...
['subplot(AxesHandle2);
[o2,o1Bound,o1Area]=drawROI(image1,o1Bound,o1Area);title(text_ROI);']);
text_eq='Enhancement by histogram Equalization';
'Position',[0,300,80,20],...
'String','his-Equalization',...
'Callback',...
['subplot(AxesHandle2);image1=histeq(uint8(image1));imagesc(image1);title(text_eq);']);
text21='Adaptive Binarization after FFT';
'Position',[0,260,80,20],...
'String','Binarization',...
'Callback',...
[%'W=inputdlg(text);W=str2num(char(W));'...
'subplot(AxesHandle1);'...
'image1=adaptiveThres(double(image1),32);title(text21);']);
text='Please input the FFT factor(0~1)';
text_fft='Enhancement by FFT';
'Position',[0,280,80,20],...
'String','fft',...
'Callback',...
['W=inputdlg(text);W=str2double(char(W));'...
'subplot(AxesHandle1);image1=fftenhance(image1,W);imagesc(image1);title(text_fft);']);
text31='Thinned-ridge map';
69
'Position',[0,200,80,20],...
'String','Thining',...
'Callback',...
['subplot(AxesHandle2);o1=im2double(bwmorph(o2,''thin'',Inf));imagesc(o1,
[0,1]);title(text31);']);
text41='Remove H breaks';
'Position',[0,180,80,20],...
'String','remove H breaks',...
'Callback',...
['subplot(AxesHandle2);o1=im2double(bwmorph(o1,''clean''));o1=im2double(bwmorph(o1,''
hbreak''));imagesc(o1,[0,1]);title(text41);']);
textn1='remove spike';
'Position',[0,160,80,20],...
'String','Removing spike',...
'Callback',...
['subplot(AxesHandle2);o1=im2double(bwmorph(o1,''spur''));imagesc(o1,
[0,1]);title(textn1);']);
%% locate minutia and show all those minutia
text51='Minutia';
'Position',[0,140,80,20],...
'String','Extract',...
'Callback',...
['[end_list1,branch_list1,ridgeMap1,edgeWidth]=mark_minutia(o1,o1Bound,o1Area,w);'...
'subplot(AxesHandle2);show_minutia(o1,end_list1,branch_list1);title(text51);']);
%Process for removing spurious minutia
text61='Remove spurious minutia';
'Position',[0,120,80,20],...
'String','Real Minutiae',...
'Callback',...
['[pathMap1,real_end1,real_branch1]=remove_spurious_Minutia(o1,end_list1,branch_list1,o1
Area,ridgeMap1,edgeWidth);'...
'subplot(AxesHandle1);show_minutia(o1,real_end1,real_branch1);title(text61);']);
%save template file, including the minutia position,direction,and ridge information
textSaveName='file name';
70
'Position',[0,100,80,20],...
'String','save',...
'Callback',...
['W=inputdlg(textSaveName);W=char(W);'...
'save(W,''real_end1'',''pathMap1'',''-ASCII'');']);
%invoke template file loader and do matching
h=uicontrol('Style','pushbutton',...
'String','Match',...
'Position',[0,80,80,20],...
'Callback',...
['finger1=fingerTemplateRead;'...
'finger2=fingerTemplateRead;'...
'percent_match=match_end(finger1,finger2,10);']);
The use of a GUI has hence rendered the project very effective as we
can now view the transitions from one processing stage to another. Also, the
print was being loaded into the MATLAB environment using the loadimage
function below which is sufficient in the absence of a fingerprint reader.
function image1=loadimage
% dialog for opening fingerprint files
%Final Year Project 2010-2011
%Akangbe S.O
%Computer Engineering,UNILAG
[imagefile1 , pathname]= uigetfile('*.bmp;*.BMP;*.tif;*.TIF;*.jpg','Open An Fingerprint
image');
if imagefile1 ~= 0
cd(pathname);
image1=readimage(char(imagefile1));
image1=255-double(image1);
end;
71

Fingerprint Database for Exam Authentication

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Fingerprint Database for Exam Authentication

Hochgeladen von

Copyright:

Verfügbare Formate

TABLE OF CONTENTS

Registration/Exam Enrollment Table

Minutiae/Finger Details Table

Course Table (Final)

Bifurcations Table (Final)

Terminations Table (Final)

Students Table (Final)

Finger Table (Final)

Registration Table (Final)

A fingerprint image acquired by an Optical Sensor

Minutia. (Valley is also referred as Furrow, Termination is

Verification vs. Identification

Simplified Fingerprint Recognition System

Data Model as a communication tool and database

Data Modeling Stage

Matlab Database Toolbox Stage

tens of thousands) due to small errors that have to be estimated.

1.3 PROBLEM STATEMENT

into the university. Matching will be done during the examination

1.4.1 To realize a database for minutiae storage using MATLAB and

integrated with a MATLAB algorithm for fingerprint

in the absence of a real students fingerprint

from a scanner/sensor. The database will also provide access to other

Among the other future work recommended is to obtain a program that

1.5 PROJECT SIGNIFICANCE

The significance must be considered in every proposed system to

1.5.2 Course Advisers

deposited in natural secretions from the eccrine glands present in friction

Figure2.0 A fingerprint image acquired by an Optical Sensor

2.1 BRIEF HISTORY OF FINGERPRINT

In 14th century Persia, a doctor, observed on various official

government papers that no two fingerprints were exactly alike.

method for obtaining such fingerprints. He is also credited with the

still in use today, and are sometimes referred to as Galton Details.

individuals in the civil files.

100 million persons' fingerprints, many in the form of two-finger

fingerprint interface standards are important to enable sharing records

2.2 FINGERPRINT RECOGNITION SYSTEM

The minutiae do not change naturally during a persons life.

Figure 2.2.1. Minutia. (Valley is also referred as Furrow, Termination is

Law Enforcement Forensics

Border Control/Visa Issuance

Figure 2.2.2 Verification vs. Identification

2.2.1 SYSTEM LEVEL DESIGN

Figure 2.2.3 Simplified Fingerprint Recognition System

For fingerprint acquisition, optical or semi-conduct sensors are widely used.

2.2.2 ALGORITHM LEVEL DESIGN

Figure 2.2.4 Minutia Extractor

Image Enhancement, a necessary stage for fingerprint recognition

selective process also requires the ridge frequency.

The minutiae extraction stage includes:

Thinning: Ridge thinning is used to eliminate the redundant pixels of

ridges till the ridges are just one pixel wide.

For the post processing stage, a more rigorous algorithm is developed to

Figure2.2.5 Minutia Matcher

2.3 MINUTIAE STORAGE

2.4 FINGERPRINT CLASSIFICATION

should be noted that this classification of fingerprints is essentially

classification is used for indexing in the fingerprint database.

fingerprint classification approach is based on the extraction of fingerprint

3.1 DESIGN STATEMENT

Here, minutiae points shall be saved in a MySQL database rather than in a

. The approach in doing this should