Department of Computer Science and Information Technology

Project Report

Project Title: Automatic Number Plate Recognition

Name of Candidate: Muhammad Shahbaz

Registration Number:15-UGLC-948

Name of Supervisor: Dr. Ilyas Shb.

Supervisor’s Signature: ……………….

Submission Date: …………………………………

1|Page
Abstract
Automatic Number Plate Recognition (ANPR) is an internationally recognized methodology that
is used in vehicle identification. ANPR systems allow for real time recognition of a vehicle’s
number plate. Vehicle parking is an important component within any transportation system,
whereby vehicles must be parked at every destination. Most vehicles are parked most of the time
during the day. There has been a problem with the car park entry registration process for
visitor/staff/student entering the Strathmore University. The problem involves a security guard,
the security guard has then to record by writing onto an occurrence book the vehicle details
include vehicle number plate, color, model, entry/exit timings as well as the ticket number
assigned to the vehicle. This process of writing is tedious and time consuming and is also prone
to inaccurate recordings, furthermore the backup and sharing of this vehicle information is
difficult because the data is hard copy. This dissertation has its basis on the adoption of software
solution operating on a mobile device that has ANPR capabilities to aid in vehicle identification
and vehicle registration. Methodology, the software developed implements Optical Character
Recognition (OCR) using the mobile device camera to detect and capture the vehicle number
plate.

2|Page
ACKNOWLEDGEMENT
First of all, I thank my most high almighty God who care and courage during my whole period of
time at the University of Sargodha. Also, I extend my acknowledgements to many individuals for
their contribution toward completion of the final year project. I would like to sincerely appreciate
my project supervisor Dr. Ilyas Shb. for the help and guidance during the period of the project
development. He has provided significant comments and advice on defining the topic and
proceeding with the project. His assistance in editing was very greatly appreciated. His
enthusiastic nature always encourages me. Finally, I thank my family for their love, morale
support and prayer along my study and thanks all friends and people who I have not mentioned.

3|Page
Table of Contents
Abstract.............................................................................................................................2
Acknowledgments.............................................................................................................3
Chapter One: Introduction.................................................................................................5
1.1 Background..................................................................................................................5
1.2 Problem Statement.......................................................................................................5
1.3 Problem Solution.........................................................................................................5
1.4 Scope...........................................................................................................................5
1.5 Objectives....................................................................................................................7
Chapter Two: Literature Review.......................................................................................7
2.1 Introduction.................................................................................................................7
2.2 Automatic Number Plate Recognition.........................................................................7
2.3 Image Acquisition .......................................................................................................7
2.4 Number Plate Detection ..............................................................................................8
2.5 Character Segmentation ..............................................................................................8
2.6 Character Recognition .................................................................................................8
2.7 Elements of ANPR Systems ........................................................................................9
2.7.1 Hardware Components of ANPR systems ................................................................9
2.7.2 Software Components of ANPR systems ..................................................................9
2.8 Number Plate and their Properties. ..............................................................................9
2.8.1 Material ......................................................................................................................9
2.8.2 Size .............................................................................................................................9
2.8.2 Color Categorizations ...............................................................................................10
2.9 Challenges…………………………………………………………………………….10
Chapter Three: Research Methodology ..............................................................................11
3.1 Introduction ...................................................................................................................11
3.2 System Development Methodology ..............................................................................11
3.3 Requirement gathering and analysis ..............................................................................12
3.3.1 Functional Requirement ..............................................................................................12
3.3.2 Non-Functional Requirement.......................................................................................12
Chapter Four: System Design and Architecture ...................................................................13
4.1 Introduction ....................................................................................................................13
4.2 Systems Architecture .....................................................................................................13
4.3 Administrator Application .............................................................................................14
4.4 Use Case Diagram .........................................................................................................14
4.5 System Sequence Diagram ............................................................................................15
Chapter Five: System Implementation and Testing ............................................................16
5.1 Introduction ...................................................................................................................16
5.2 System Development Tools ..........................................................................................16
5.3 System Testing ..............................................................................................................16
5.3.1 Functionality Testing ..................................................................................................16
5.3.2 Performance Testing ...................................................................................................16
5.4 Application Testing ........................................................................................................16
Chapter Six: Discussion of the Results from Testing ...........................................................18
6.1,2 Conclusion and Recommendation ...............................................................................18
6.3 Suggestion for Future Work ...........................................................................................18
4|Page
 Chapter one: Introduction

1.1. Background
Most Pakistan cities are populated and other cities are not populated. So if u see the
populated cities in Pakistan, there are not manageable parking system so we see the parked
vehicles on the road or the public places, if where parking systems develop there are more
time consuming and manual systems with some labor powers. If we develop and manage our
parking systems then of course it will be the beneficial for the people and of course for the
government.

1.2. Problem Statement

With the rising number of vehicles, it is getting difficult to effectively manage vehicles. The
vital question here is how to find a particular vehicle? The obvious response to this question
is by using the vehicle’s number plate. A security guard having to confirm membership
details by checking for membership sticker on the windscreen of the vehicle or by checking
the driver’s identification card. The security guard has then to record by writing onto an
occurrence book the vehicle details such as vehicle number plate, color, model, entry/exit
timings as well as the ticket number assigned to the vehicle. This process of writing is time
consuming and is also prone to inaccurate recordings, furthermore Manual Systems also
contain the IR id or some specific tag or barcode which work correctly but its for only the
staff or office tagged vehicles. So there are some problem occur.

1.3. Problem solution

The proposed is automated prepaid system to detect, the number plate detection on entry
basis or also detect from exit basis. Its all work done through HD camera. And no tag or
RFID need to detect the specific barcode but License plate

1.4. Scope
License plate recognition system consists of four major processes namely: Image acquisition,
Number plate localization Character segmentation and recognition. This study only focused
on these four areas in relation to image processing techniques that are required in the number
plate recognition process for the automation of the vehicle entry registration process.

5|Page
 1.5. Objectives
There are following objectives:

 To design and implement a number plate recognition system for capturing and
convert image to variable digits.
 Pre-process the images
 Detect the LP
 Recognize the Number placed on LP.

6|Page
 Chapter Two: Literature Review

2.1 Introduction
According to Pakistan the registered vehicles in the country increased by 9.6 per cent in 2018 as
the number of vehicles have reached 23,588,268 in the last year compared to 21,506,641 vehicles
in 2017, according to the data of Pakistan Bureau of Statistics (PBS) ISLAMABAD. the previous
year. With the increase number of vehicles its necessary to have parking area access with slow
time consuming.

2.2 Automatic Number Plate Recognition

Automatic number plate recognition (ANPR) is a surveillance method that was first developed in
1976 at the Police Scientific Development Branch in the United Kingdom with aim of combating
crime (Cornwall, 2009). Automatic Number Plate Recognition is a process where vehicles are
identified or recognized using their number plate or license plate. ANPR uses image processing
techniques in order to extract the vehicle number plate from digital images.
ANPR systems normally comprises of two components: A camera that used in capturing of
vehicle number plate images, and software that extracts the number plates from the captured
images by using a character recognition tool that allows for pixels to be translated into numerical
readable characters. A license plate recognition system generally works in four main parts
namely image acquisition, license plate recognition/detection, characters segmentation, and lastly
character recognition. Figure 2.2 shows typical ANPR process that result in text from an input
image.

Figure 2.2 Typical ANPR Process

2.3 Image Acquisition

The first step is the image acquisition stage the image of the vehicle is captured using a
photographic camera. The constraint is that the image of the vehicle should be captured in such a
way that the selected input image contains rear or front view of the vehicle with the number plate
The image is usually captured in a RGB (Red, Green and Blue) color model. The captured image

7|Page
is affected by many factors like:
 optical system,
 distortion,
 system noise,
 lack of exposure or excessive relative motion of camera or
RGB image into gray color, noise removal, and border enhancement for brightness. Image pre –
processing is usually done through image filtering. The image acquisition stage heavily depends
on the camera. Proper camera setup is essential for accurate deployment of the ANPR system the
properties of the camera form an important part in the image acquisition process. A blurry image
is a factor that greatly affects the accuracy of the ANPR system. The positioning of a permanent
camera is a factor affecting the performance of ANPR systems, whereby camera positioning
needs to be carefully surveyed.

2.4 Number Plate Detection

The next stage that follows is the license plate recognition/number plate recognition phase that
does several functions such as resizing of the image to a feasible aspect ratio. As well as
converting the colored image into a grey scale image. Number plate detection searches an input
image in order to identify specific features that contain the number plate. The number plate can
be found anywhere within an image, it is impractical to check all the pixels of the image in order
to locate the number plate. Therefore, we only focus on those pixels that have the number plate.
Number plate detection methods can be categorized into three categories: color-based, edge-
based, and texture-based.

2.5 Character Segmentation

Character segmentation can be defined as a technique, which partitions images of lines or words
into individual characters. It is an operation that seeks to decompose an image of a sequence of
character into sub images of individual symbols (Kumar & Singh 2011). Character segmentation
is an operation that seeks to decompose an image of a sequence of characters into sub-images of
individual symbols.
Character segmentation is the process through which the text component within an image is
isolated from the background. In order for proper text recognition to take place the line of text is
first segmented, then from the segmented line the words are segmented and then from that the
characters are segmented.

2.6 Character Recognition

Character recognition is process of detecting and recognizing characters from input image and
converting it into meaningful text in ASCII (American Standard Code for Information
Interchange) or other equivalent machine editable form. Character recognition is the process to
classify the input character according to the predefined character class. Template matching, or
matrix matching, is one of the most common classification methods.

8|Page
In template matching, individual image pixels are used as features. Classification is performed
by comparing an input character image with a set of templates from each character class.

2.7 Elements of ANPR Systems

Automatic Number Plate Recognition Systems normally consist of hardware and software units.

2.7.1 Hardware Components of ANPR systems include:

 Camera(s) – Digital cameras are used in the image acquisition stage
 Infra-Red – Light source at night illumination to camera
 Frame Grabber – This is a hardware interface
 Computer – This is the central processing unit in ANPR systems
2.7.2 Software Components of ANPR systems include:
 ALPR Software – This software has the OCR capabilities which results in the
extraction of the number plate from the digital image.
 Database – This provides storage for the data (number plates that have been read
by the ANPR).
 Back End Software - The back end software is located on a server and provides
many functions such as

2.8 Number Plate and their Properties.

Number plate identification on material and size required so Number plate is categorized by the
following factors;

2.8.1 Material:
Aluminum is the most commonly materials used to manufacture number plates in Tanzania
because the material possesses great capacity to reflect infrared radiation hence due to its
infrared reflector qualities make it the ideal metal for number plate manufacturing worldwide.

2.8.2 Size;
6-inch height and 12 inches

9|Page
2.8.3 Color categorizations;

2.9 Challenges
There are a number of factors that may cause the ANPR system not to function effectively some
of these factors include: Misidentification: In case the number is read partially, the remote
computer might identify the number plate incorrectly or would not be able to decrypt at all.
Characters and numbers may be detected wrongly and interchanged identity may occur;
characters may be read as numbers and vice versa. Problems in recognition under bad weather
conditions: Natural Calamities like heavy rainfall or snowfall may hinder the performance of the
detection system. Non-standardized number
There are a number of possible difficulties that the software must be able to cope with. These
include:
 Poor image resolution, usually because the plate is too far away but sometimes resulting
from the use of a low-quality camera; ii. Blurry images, particularly motion blur;
 Poor lighting and low contrast due to over exposure, reflection or shadows;
 An object obscuring (part of) the plate, quite often a tow bar, or dirt on the plate;
While some of these problems can be corrected within the software, it is primarily left to the
hardware side of the system to work out solutions to these difficulties. Proper camera position
could help overcome some of these challenges.

10 | P a g e
 Chapter Three: Research Methodology

3.1 Introduction
Research design is the overall plan for connecting the conceptual research problems to the
pertinent (and achievable) empirical research. In other words, the research design articulates
what data is required, what methods are going to be used to collect and analyze this data, and
how all of this is going to answer your research question.
This chapter describes the methods adopted in this research and explains why certain methods
were preferred over others. It also highlights a number of issues relating to the research process,
research facilitation and field-based research within a school environment.

3.2 System Development Methodology

This research adopted an Object-oriented System Development methodology. An Object-
oriented
software design is a design strategy where system designers think in terms of ‘things’ instead of
operations or functions. The executing system is made up of interacting objects that maintain
their own local state and provide operations on that state information. Object-oriented analysis
and design is more cost-effective and a faster way to develop software and systems. This
technology cuts development time, overhead and enables software engineers to make reusable,
reliable and easily maintainable applications
Requirement analysis  Design  Code Test  Maintenance. Following Fig 3.2 elaborate
the development cycle.

Figure 3.2 Development Process

11 | P a g e
3.3 Requirements gathering and analysis
Requirements are gathering from the related persons according to project, implementation of a
pre-paid smart recognition of vehicles, understanding the operating environment, reading
literatures and talking with employers of different relevant workplace so as to get the
requirements of the system . The system to be designed must meet both functional and non-
functional requirements. The complete system consists of hardware and software components
built together to meet the functionality of the main objective

3.3.1 Functional requirements

The following are the system functional requirements of the system;
 The system should be able to capturing the vehicles.
 The system should be able to identify the LP of vehicle.
 The system should be able to detect and recognize the vehicle number plate

3.3.2 Non-functional requirements

Non-functional requirements define the overall qualities or attributes of the resulting system.
Non-functional requirements are constraints of the product developed that must meet to make the
system useful. These includes the followings.
 Performance – The system use short response time, high throughput and easy Utilization.

 Accuracy- System should work accurate.

 Scalability- The system can be extended to increase total throughput under an increased
load when resources (typically hardware) are added.

 Availability- the system is available for service when requested by the user at any time.

 Reliability- the system performs its required functions under stated conditions for a
specific period of time. Fault free

 Maintainability- the system is maintainable to incorporate other functionalities.

 Security- Unauthorized access to the system is not allowed.

12 | P a g e
 Chapter Four: System Design and Architecture

4.1 Introduction
Modeling is the initial step undertaken during the process of system design. Modeling provides
guidance on how to develop the system.

4.2 System Architecture

A system architecture or systems architecture is the conceptual model that defines the structure,
behavior and views of a system. The Figure 4.2 is an architectural design for the Automatic
Number Plate Capture System.

Figure 4.2 Architecture OF ANPR

13 | P a g e
4.3 Administrator Application
This comprises the admin side that has the Optical Character Recognition functionality
required in the scanning of vehicle number plate, as well as inputting other vehicle description
details. The admin application is communicating with the server side via wireless local area
network (WLAN) after getting the image from the camera and recognize the image.

4.4 Use Case Diagram

A use case diagram at its simplest is a representation of a user's interaction with the system and
depicting the specifications of a use case (Use Case Diagram). A use case diagram can portray
the different types of users of a system and the case and will often be accompanied by other
types
of diagrams as well. The Figure 4.4 is a use case diagram describing the actors and the activities
involved in the new vehicle registration process.

Sensor

Capture Camera
Image

Processing Find Plate

Region

Recognition Segmentation Plate

Extractio
n

Figure 4.4 Use Case Diagram Of ANPR

14 | P a g e
4.5 System Sequence Diagram
System sequence diagrams are visual summaries of a use case. The Figure 4.5 is a System
Sequence Diagram that depicts the add vehicle details use case. This use case has a pre-condition
that requires the scan number plate process to have occurred successfully.

Figure 4.5 Sequence Diagram of ANPR

15 | P a g e
 Chapter Five: System Implementation and Testing

5.1 Introduction
This chapter we discuss the development and the testing of the system. While using an Object-
Oriented System Development methodology the researcher was able to test functional units of
project.

5.2 System Development Tools and Languages.

With regards to the system architecture discussed in the previous chapter. The tools required in
the project are:
 Python
 OpenCV
 AI algorithm
 Matplotlib
 NumPy
 Pandas
 Jupyter Note Book Or PyCharm IDE

5.3 System Testing

The System Testing methods that the researcher has adopted due to the Object Oriented
Development Methodology are described next:

5.3.1 Functionality Testing

Functionality testing is performed to verify that a software application performs and functions
correctly according to design specifications. During functionality testing we check the core
application functions , like capturing, detecting LP etc. Functionality testing verifies that
an application is still fully functional.

5.3.2 Performance Testing

Performance testing is helpful in developing higher quality software in less time while reducing
development costs. The goal is to test performance early and often in the development effort.
Software performance testing mean here, accuracy of recognizing the LP is better and fault free
or not accurate according to desire.

5.4 Application Testing

The first release of the application was tested by the actual users of the system, the users
primarily comprised of the Security Guards at the different entry points of the parking place.
performed in order to get the users feedback on how the actual system was working in the real

16 | P a g e
world. The first test that was carried out was a functional test that was testing the most complex
and challenging part of this system which included the use of Optical Character Recognition
processes
(OCR) in order to detect and display the number plate in a plain text format. This process is also
referred to as scanning of the vehicle number plate. This process crucial in this system and it
forms the basis of the entire system. The results of the unit test are recorded in figure 5.4a and
5.4b.

Figure 5.4a

17 | P a g e
Figure 5.4b
 Chapter Six: Conclusions and Recommendations

This dissertation was action oriented. It sought to implement a solution to the challenges faced
by the Security personnel in terms of vehicle identification and vehicle registration for all the
vehicles that enter and exit the parking premises. The result was the implementation of software
application operating on a mobile device that allows for easy identification and registration of
vehicles.

6.1 Conclusions
The researcher has been able to successfully implement the use of Optical Character Recognition
(OCR) that allows for the automatic detection, recognition and display of
vehicle number plates. The OCR process forms the basis of the entire system that was proposed
by the researcher.
Some of the challenges of this project that, the ability of the system to detect a vehicle is
irrespective of the weather condition like excessive rain or heat and irrespective of the area of
location within a specified distance.

6.2 Recommendations
An Automatic Number Plate Recognition System is primarily a tool for vehicle identification.
With proper implementation an ANPR system can be useful in:
 Identification of vehicles that have been reported as stolen.
 Identification of vehicles that are wanted for having committed traffic offenses.
 Identification of vehicles that have fake registration certificates.

6.3 Suggestion for future work

ANPR technologies can be implemented in several scenarios whereby there is a need to monitor
all vehicles that are accessing or exiting from a particular vicinity. For example ANPR
technology can greatly enhance border patrol and border surveillance whereby all the vehicles
entering or exiting the country can be quickly checked to see if there are any known offences that
relates to hat particular vehicle before the vehicle is granted entry or exit into the country.

18 | P a g e