AUTOMATED NUMBER PLATE RECOGNITION SYSTEM FOR EFFICIENT ENTRY

AND EXIT MANAGEMENT AT INSTITUTIONAL ENTRANCES

PROJECT REPORT

Submitted in partial fulfillment of the requirements for the award of a degree of

BACHELOR OF TECHNOLOGY

In

COMPUTER SCIENCE AND ENGINEERING BY

P. YESWANTH S. SANDEEP

(20331A05D3) (20331A05F9)

M. SUDHEER P. GAYATHRI

(20331A05J1) (21335A0514)

Under the Supervision of

Mr.P.Ram Santosh Naidu

Assistant Professor
Maharaj Vijayaram Gajapathi Raj (MVGR) College of Engineering (Autonomous)
Vizianagaram

CERTIFICATE

This is to certify that the project report entitled ― Automated Number Plate Recognition
System for Efficient Entry and Exit Management at Institutional Entrances being submitted
by P.Yeswanth, S.Sandeep, M.Sudheer, P.Gayathri bearing registered numbers 20331A05D3,
20331A05F9, 20331A05J1, 21335A0514 respectively, in partial fulfillment for the award of the
degree of ―Bachelor of Technology” in Computer Science and Engineering is a record of
bonafide work done by them under my supervision during the academic year 2022-2023.

Supervisor HOD CSE

Mr.P.Ram Santhosh Naidu Dr. P. Ravi Kiran Varma
Assistant Professor, Head of the Department,
Department of CSE, Department of CSE,
MVGR College of Engineering, MVGR College of Engineering,
Vizianagaram. Vizianagaram.
 DECLARATION

We hereby declare that the work done on the dissertation entitled “Automated Number Plate
Recognition System for Efficient Entry and Exit Management at Institutional Entrances”
has been carried out by us and submitted in partial fulfillment for the award of credits in Bachelor
of Technology in Computer Science and Engineering of MVGR College of Engineering
(Autonomous) and affiliated to Jawaharlal Nehru Technological University (Kakinada). The
various contents incorporated in the dissertation have not been submitted for the award of any
degree from any other institution or university.
 ACKNOWLEDGEMENTS

We would like to express our sincere gratitude to our project guide Mr. Ram Santhosh
Naidu for his valuable suggestions and whole-hearted guidance in successfully completing
our project and we are extremely grateful for his constant encouragement and continuous
support throughout this work.
We also thank Dr. K. V. L. Raju, Principal, and Dr. P. Ravi Kiran Varma, Head of the
Department, for extending their utmost support and cooperation in providing all the provisions
for the successful completion of the project. We sincerely thank all the members of the staff in
the Department of Computer Science & Engineering for their sustained help in our pursuits.
We thank all those who contributed directly or indirectly in successfully carrying out this
work.
 ABSTRACT
Efficient entrance and exit management have become one of the biggest concerns for any
organization. we notice a manual description of any vehicle that enters and exits, which is
observed by the particular people and scripted in the books. That is stored and further used for
other purposes. which can be time-consuming, paper wastage, and prone to errors. So,
automation of such safety and security is essential.

Modern technologies such as electronic access control can help to overcome this limitation by
providing more secure authentication methods. Automatic number plate recognition technology
can help by identifying vehicles quickly and accurately. ANPR cameras and software can detect
and recognize license plate numbers, monitor traffic flow, track vehicle movements, and enforce
parking regulations. ANPR technology can enhance security by detecting unauthorized vehicles
and preventing vehicle theft. It can flag suspicious vehicles, alert security personnel, and trigger
alarm systems to maintain an accurate record of vehicle movements.
 TABLE OF CONTENTS

Topic Page Number

List of Abbreviations
Introduction
Objectives, Scope, Limitations
Literature Survey
Procedural Design
Flow Chart
Algorithm
 LIST OF ABBREVIATIONS

AI - Artificial Intelligence

ML - Machine Learning

OpenCV - Open-Source Computer Vision

ROI - Region of Interest

HOG - Histogram of Oriented Gradients

SVM - Support Vector Machine

CNN - Convolutional Neural Network

YOLO - You Only Look Once

BBox - Bounding Box

OCR - Optical Character Recognition

ANPR - Automatic Number Plate Recognition

LPR - License Plate Recognition

 INTRODUCTION

As we can observe that at the entrance of any institutions, malls, or any other authorized
buildings we notice a manual description of any vehicle that enters and exits. Which is observed
by the particular people and scripted in the books. that is stored and further used for other
purposes.

Our motto is to introduce a system that swaps the manual script written by people with
computerized data. which is easy to maintain and not a big deal of human effort which needs a
needs machine learning procedure. using image processing, we will detect the vehicle's number
plate in the captured picture.

As we know the identity of a vehicle is recognized by its number plate. So here we are trying to
identify numbers from the number plate through the captured images. And make further
processing on it. To particularly get the data from the considered numbers. Here we capture an
image of the vehicle and identify its number plate we can retrieve the data from the system when
the vehicle checks in and check out. This makes it easier to collect information and reduces the
interruption to vehicle users and those who gather information manually.
Objectives:
1. Automatic Number Plate Recognition (ANPR) is an image processing technology that uses a
number (license) plate to identify the vehicle. The objective is to design an efficient automatic
authorized vehicle identification system by using the vehicle number plate.
2. The sole purpose of the automated system is to distinguish vehicles by recognizing number
plates. This number plate recognition process is divided into three parts:
 Identification of vehicle number plate
 Plate character recognition and segmentation
 OCR calculation to translate characters into encoded content.
3. The goal of speed detection is to develop various methods that government agencies can
implement. The most common methods include RADAR and LIDAR devices to detect the
speed of the vehicles. A RADAR device sends a radio signal to a moving object the signal
reflected is then received by a receiver

Scope:
1. This system uses image processing techniques for the recognition of the vehicle from the
database stored in the computer. The system works satisfactorily for a wide variety of
conditions and different types of number plates.
2. The ANPR market will grow by around 25-30 percent in the next five years. Enforcement
solutions combining, for example, speed measurement, camera, and ANPR will drive future
growth.
3. The future scope is that the automatic vehicle recognition system plays a major role in
detecting threats to defense Also it can improve security related to women as they can easily
detect the number plate before using a cab or other services.

Limitations:
1. A disadvantage of ANPR parking systems is that they rarely take into account human error
and behavior. ANPR systems do not usually consider giving a grace period when you enter a
car park
2. Firstly, the images of the number plate or of any object which is taken by using the optical
character reader technology may get blurred mainly due to the reason of motion blurring for
which the picture seems to be hazy when uploaded in the database.
3. Secondly, the technology often uses low-resolution images for which the images are not
 actually visible properly in every case.

LITERATURE SURVEY

1. Automatic Number Plate Recognition (ANPR) is a fairly well explored problem with many
successful solutions. However, these solutions are typically tuned towards a particular
environment due to the variations in the features of number plates across the world.
Algorithms written for number plate recognition are based on these features and so a
universal solution would be difficult to realize as the image analysis techniques that are used
to build these algorithms cannot themselves boast hundred percent accuracy. The focus of
this paper is a proposed algorithm that is optimized to work with Ghanaian vehicle number
plates. The algorithm, written in C++ with the OpenCV library, uses edge detection and
plate. The Tesseract OCR engine was then used to identify the detected
characters on the plate.
2. Due to a huge number of vehicles, modern cities need to establish effectively automatic
systems for traffic management and scheduling. One of the most useful systems is the
Vehicle License-Plate (VLP) Recognition System which captures images of vehicles and
read these plates’ registration numbers automatically. In this paper, we present an automatic
VLP Recognition System, ISeeCarRecognizer, to read Vietnamese VLPs’ registration
numbers at traffic tolls. Our system consists of three main modules: VLP detection, plate
number segmentation, and plate number recognition. In VLP detection module, we propose
an efficient boundary line-based method combining the Hough transform and Contour
algorithm. This method optimizes speed and accuracy in processing images taken from
various positions. Then, we use horizontal and vertical projection to separate plate numbers
in VLP segmentation module. Finally, each plate number will be recognized by OCR module
implemented by Hidden Markov Model. The system was evaluated in two empirical image
sets and has proved its effectiveness (see section IV) which is applicable in real traffic toll
systems. The system can also be applied to some other types of VLPs with minor changes.
3. This paper presents a method for recognition of the vehicle number plate from the image
using neural nets and mathematical morphology. The main theme is to use different
morphological operations in such a way so that the number plate of the vehicle can be
extracted efficiently. The method makes the extraction of the plate independent of color, size
and location of number plate. The proposed approach can be divided into simple processes,
which are, image enhancement, morphing transformation, morphological gradient,
combination of resultant images and extracting the number plate from the objects that are left
in the image. Then segmentation is applied to recognize the plate using neural network. This
 algorithm can quickly and correctly recognize the number plate from the vehicle image

PROCEDURAL DESIGN:

ENTRY System:
Initially, the execution starts by capturing the vehicle by camera and Vehicle Number Plate is
extracted from the captured image. From the Vehicle Number Plate image, number image alone
extracted and it is converted into text. Whenever the captured number plate matched with the
database number plates images then it comes under the organizational vehicles. Then the matched
number plate is stored in the database along with the entry time of the vehicle at the entry gate.
On condition that, number plate is not go with the number plates stored in the database, then that
number plate will be categorized under Outsider’s Vehicle. Hence entry time alone is noted.
Finally, the entry system execution stopped.

EXIT System:
The same process is pursued during the initial stages like Acquisition, pre-processing and
converting image to text. When this process is completed, the exit time (time of capturing the
image of the vehicle at the exit gate) is saved. Then this exit execution stopped.

Flow Chart:
Algorithm:

Here we used Haar Cascade, an algorithm that can detect objects in images, irrespective of their
scale in image and location. This algorithm is not so complex and can run in real time. We can
train a haar-cascade detector to detect various objects like cars, bikes, buildings, fruits, etc.

Haar Cascading:
Haar classifiers were classifiers that were used in the first real-time face detector. A Haar
classifier, or a Haar cascade classifier, is a machine-learning object detection program that
identifies objects in an image and video.
Making a Haar Cascade Classifier
The algorithm can be explained in four stages:
 Calculating Haar Features
 Creating Integral Images
 Using Adaboost
 Implementing Cascading Classifiers
It’s important to remember that this algorithm requires a lot of positive images of faces
and negative images of non-faces to train the classifier, similar to other machine learning models.

Calculating Haar Features

The first step is to collect the Haar features. A Haar feature is essentially calculations that are
performed on adjacent rectangular regions at a specific location in a detection window. The
calculation involves summing the pixel intensities in each region and calculating the differences
between the sums. Here are some examples of Haar features below.

Types of Haar features:

These features can be difficult to determine for a large image. This is where integral images come
into play because the number of operations is reduced using the integral image.
Creating Integral Images
Without going into too much of the mathematics behind it (check out the paper if you’re interested
in that), integral images essentially speed up the calculation of these Haar features. Instead of
computing at every pixel, it instead creates sub-rectangles and creates array references for each of
those sub-rectangles. These are then used to compute the Haar features.

Adaboost Training

Adaboost essentially chooses the best features and trains the classifiers to use them. It uses a
combination of “weak classifiers” to create a “strong classifier” that the algorithm can use to
detect objects.

Weak learners are created by moving a window over the input image, and computing Haar
features for each subsection of the image. This difference is compared to a learned threshold that
separates non-objects from objects. Because these are “weak classifiers,” a large number of Haar
features is needed for accuracy to form a strong classifier.

Representation of a boosting algorithm.

The last step combines these weak learners into strong learners using cascading classifiers.
Implementing Cascading Classifiers

A flowchart of cascade classifiers.

The cascade classifier is made up of a series of stages, where each stage is a collection of weak
learners. Weak learners are trained using boosting, which allows for a highly accurate classifier
from the mean prediction of all weak learners.
Based on this prediction, the classifier either decides to indicate an object was found (positive) or
moves on to the next region (negative). Stages are designed to reject negative samples as fast as
possible because a majority of the windows do not contain anything of interest.
It’s important to maximize a low false negative rate because classifying an object as a non-object
will severely impair your object detection algorithm. The video below shows Haar cascades in
action. The red boxes denote “positives” from the weak learners.
Haar cascades are one of many algorithms that are currently being used for object detection. One
thing to note about Haar cascades is that it is very important to reduce the false negative rate, so
make sure to tune hyperparameters accordingly when training your model

The applications of this technology are enormous in a variety of different fields. A few of the
most important applications are listed below:
 Facial Recognition
 Robotics
 Autonomous Vehicles
 Image Search and Object Recognition
 Agriculture
 Industrial Use
 DATA COLLECTION

INPUT:
The Video captured by the Camera are inputs
OUTPUT:
The Vehicle Number plate is detected and stored in a database
SOURCES OF DATA:
https://github.com/topics/yolov4-tiny

HARDWARE PROCUREMENT

1. SSD : 512GB
2. RAM : 16GB
3. Processor : Intel Core[i5]

SOFTWARE PROCUREMENT

1. Operating System : Windows 11

2. Coding Language : Python
3. Tools : Visual Studio Code, MySQL Workbench, Webcam
WORKING PRINCIPLE:

Image Acquisition:
The camera is used to capture the image of the vehicle at the entrance. The webcam is connected
to the processor of the system. The ultrasonic sensor gives a trigger pulse to the camera whenever
it detects the vehicle and the camera captures the image and sends it to the processor.

Image Pre-Processing:
Image preprocessing is followed to eradicate the added noise and distortion present in the
captured image. Here we use a Gaussian filter and adaptive thresholding to preprocess the image.

Localization of Number Plate:

The preprocessed image is applied to the edge operator in order to extract the license plate region
from the background. Generally, the license plate is in rectangular shape, this region of interest is
obtained by comparing it with the different dimensionality of the edge. The contour detection
method is followed, which extracts all the closed shapes or boundaries from the image.

Character Segmentation:
In the optical character recognition (OCR) system, Python-tesseract is an important tool used in
Python. It will read and recognize the text information present in the images. If it is used as a
script then it will print the recognized text instead of writing it to a file The character is using the
technology of Optical Character Recognition (OCR) which extracts the characters present
in the license plate.
CONCLUSION:
Vehicle tracking is important for traffic control, surveillance, and finding stolen cars. Real-time
license plate detection and recognition are crucial, but challenging due to variations in
background, font color, style, size, and non-standard characters. To improve efficiency, a deep-
learning strategy was used. Images were captured under various conditions and validated for high
recognition rates. This approach can benefit law enforcement and private organizations for
homeland security. Future work may include training and validation using hybrid classifiers and
improving recognition in varying weather conditions.

To allow the feasibility of maintaining a socially able interface with the users in the case of the
vehicle detecting and making a log of it. So that it should be easy to grab information about the
particular vehicle.

Indeed, this technology provides an easy-to-understand model, cost-effective, better, faster and
touchless vehicular identification.

From automating tasks to managing space, improving the mobility of people and goods,
reducing traffic congestion and managing incidents effectively.
References:

1. Patel, Chirag, Dipti Shah, and Atul Patel. "Automatic number plate recognition system
(ANPR): A survey." International Journal of Computer Applications 69.9 (2013).
2. 2 Puranic, Aniruddh, K. T. Deepak, and V. Umadevi. "Vehicle number plate recognition
system: a literature review and implementation using template matching." International
Journal of Computer Applications 134.1 (2016): 12-16.
3. Qadri, Muhammad Tahir, and Muhammad Asif. "Automatic number plate recognition
system for vehicle identification using optical character recognition." 2009 International
Conference on Education Technology and Computer. IEEE, 2009.
4. https://www.irjmets.com/uploadedfiles/paper//issue_7_july_2022/28257/final/
fin_irjmets1657893552.pdf

TEAM MEMBERS:

 P.YESWANTH
 S.SANDEEP
 M.SUDHEER
 P.GAYATHRI

SIGNATURE OF THE SUPERVISOR

Mr.P.Ram Santosh Naidu