Design and Traffic Ammercement

VISVESVARAYA TECHNOLOGICAL UNIVERSITY
BELGAVI-590014
A Project Report
On
Traffic Amercement Automation
A Project report submitted in partial fulfillment of the requirements for the VIII Semester degree of
Bachelor of Engineering in Computer Science and Engineering
of Visvesvaraya Technological University, Belagavi.
Submitted by
Chirag Suthar 1DT19CS033

Chirantan Banerjee 1DT19CS034
Gaurav Mourya 1DT19CS047
Ishan Makharia 1DT19CS057
Under the Guidance of:

Mrs. Arpitha Vasudev
Asst. Professor
Computer Science and Engineering
Dayananda Sagar Academy of Technology & Management
Department of Computer Science and Engineering

DAYANANDA SAGAR ACADEMY OF TECHNOLOGY AND MANAGEMENT
Udayapura, Kanakapura Road, Bangalore-560082
2022-2023
DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING
CERTIFICATE
Certified the project work entitled Traffic Amercement Automation carried outby Chirag Suthar
bearing USN 1DT19CS033 ,Chirantan Banerjee bearing USN 1DT19CS034 , Gaurav Mourya
bearing USN 1DT19CS047 and Ishan Makharia bearing USN 1DT19CS057 bonafide students of
Dayananda Sagar Academy of Technology and Management in partial fulfilment for the
award of Bachelor of Engineering in Computer Scienceand Engineering of the Visvesvaraya
Technological University, Belagavi during the year 2022-23.It is certified that all
corrections/suggestions indicated for Internal Assessment have been incorporated in the Report
deposited in the departmental library. The project report has been approved as it satisfies the
academic requirements in respect of Project work prescribed for the said Degree.
Mrs. Arpitha Vasudev Dr. Kavitha C Dr. M. Ravishankar

Asst. Professor Professor & HOD, Principal
Dept. of CSE, DSATM Dept. of CSE, DSATM DSATM
External Examiners
Sl. No. Name Signature with Date
1.
2.
DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING
DECLARATION
We, Chirag Suthar [1DT19CS033], Chirantan Banerjee [1DT19CS034], Gaurav Mourya

[USN19CS047] and Ishan Makharia [1DT19CS057], students of VIII Semester B.E., in
Computer Science and Engineering, Dayananda Sagar Academy of Technology & Management
hereby declare that the Project entitled “Traffic Amercement Automation” has been carried out
by us and submitted in partial fulfillment of the requirements for the VIII Semester degree of
Bachelor of Engineering in Computer Science and Engineering of Visvesvaraya Technological
University, Belagavi during academic year 2022-2023.
Sl. No. Name USN Signature with date

1. Chirag Suthar 1DT19CS033
2. Chirantan Banerjee 1DT19CS034
3. Gaurav Mourya 1DT19CS047
4. Ishan Makharia 1DT19CS057
ACKNOWLEDGEMENT
We are thankful to Dr. M. Ravishankar, Principal, DSATM, Bangalore, for providing the
necessary facilities and for the encouragement to carry out the Project in DSATM.
We express our humble gratitude and thanks to Dr. Kavitha C, Professor and HOD, Computer
Science and Engineering, DSATM for guiding us and having facilitated us to complete the Project
successfully.
We are very thankful to our guide Mrs. Arpitha Vasudev, Asst. Professor, Department of
Computer Science and Engineering, DSATM for her/his constant encouragement, guidance and
support throughout the course of the Project without whose efforts this would not have been
possible.
We finally thank to all the Technical Staff for all the support and help while carrying out the project.
Last but not the least we wish to thank our friends and parents who have been encouraging us
constantly and inspiring throughout the execution of the project.
Name USN
Chirag Suthar 1DT19CS033
Chirantan Banerjee 1DT19CS034
Gaurav Mourya 1DT19CS047
Ishan Makharia 1DT19CS057
ABSTRACT
Bribery is one of the major issue in world.It is one of the pillar in corruption ecosystem.Bribes
are often demanded in situations where road users have committed an offence such as
speeding, overloading, or driving unlicensed or un-roadworthy vehicles. Bribery in these
instances may be used to ensure that the offender escapes a stiffer penalty (i.e. a R100 bribe is
requested when the alternative is to pay a legitimate fine of double that amount). Government
has acknowledged the existence of corruption and the threat that this pose towards road safety.
The Road to Safety Strategy puts a high priority on co- ordinated action in all the key areas of
road safety: law compliance, driver fitness and training, vehicle condition, fleet regulation,
overload control, pedestrian education, direct community road safety participation and active
passengers and public empowerment.
Through this project we want to automate the process of taking penalty charges in case of any
traffic violation thereby eliminating the need of any middlemen such as police. Our system
automatically detect any traffic violation and gets the info details about the vehicle’s owner,
then we generate and send a payment request to the owner, If the owner pays the amount then
it is well and good else after a few time period owner’s license and every other document
related to the vehicle will be declared null and void .
TABLE OF CONTENTS
Chapter No. Page

no.
Abstract
1 Introduction 8
2 Literature Survey 10
3 Problem statement & proposed System 15
4 High Level Design 18
5 Low level Design 23
6 Implementation 26
7 System Testing 30
8 Conclusion 35
9 Future scope 36
10 Snapshots 37
11 References 39
12 Copy of the Published paper Project Phase 1
13 Copy of the Published paper Project Phase 2

LIST OF FIGURES
Figure No. FIGURE CAPTION Page No.

4.1 Data Flow Diagram Level 0 19
4.2 Data Flow Diagram Level 1 20
5.1 Methodology Flow Diagram 25
6.1 Fine req is send to users phone number 27
6.2 Fine req is send to users email 28
6.3 E-challan website 28
7.1 Unpaid fine 34
7.2 Paid fine 34
10.1 QR code generation 37
10.2 Unpaid or created fine 37
10.3 E-challan 38
TrafficAmercementAutomation
CHAPTER 1
INTRODUCTION
Overview
Project aims to design an efficient and effective model for medical image analysis by
classifying Magnetic Resonance Imaging (MRI) based on the presence of brain tumors. The
proposed method can extract robust features from MRI images, locate and classify tumors, and
may serve as a replacement to existing manual brain tumor classification methods. Radiologists
and doctors stand to benefit from this project.
1.1 Computer Vision

Computer vision is the field of computer science that focuses on replicating parts of the
complexity of the human vision system and enabling computers to identify and process
objects in images and videos in the same way that humans do. Until recently, computer vision
only worked in limited capacity. Thanks to advances in artificial intelligence and innovations
in deep learning and neural networks, the field has been able to take great leaps in recent
years and has been able to surpass humans in some tasks related to detecting and labeling
objects. In this project, we are going to develop a detection system using computer vision,
which basically captures a QR code along with number plate attach to a car. If the driver
breaks certain traffic rule then our system scan that QR code and automatically initiates a G-
pay request to the car-owner mobile number then the owner has to pay the fine.
1.2 Objectives
The main objective of this qr detection and amercement is to provide:
 It will strictly upholds the traffic rules and discipline.
 It will greatly reduce number of accidents which mostly occur due to absence of helmet
if it’s a two wheel vehicle or over speeding of vehicle or both.
 It will greatly reduce or eliminate any corruption in the system like bribery as most of
the people can get away with their by simply bribing the police.
 This will minimize the risk factor of candidates leaving the company even after
receiving the offer letter and this will save the companies from such losses.
Dept. Of CSE, DSATM 2022-23 Page 8

 The companies will also have a credibility score so if they revoke the offer of any
candidate without proper justification their score will be impacted which will be a bad
impression for them.
1.3 Problem statement

Vehicle and Plate Detection : In general, a camera captures the vehicle's pictures as well as
the number plates and also the laptop processes the captured images and detects the car's
distance relative to the camera. Finally it reads the data by applying numerous image
processes and optical character recognition techniques.
Road and Traffic Conditions : Road conditions are a lot of varied and traffic is unstructured,
there’s lack of discipline and over loading of vehicles and their movement is kind of natural.
Recognition System Limitations : During the recognition system procedure, there exists lots
of obstruction. For example, camera limitation and natural scene effects are the main reason
to get a good image .
Fine Collection Automation : Using the number plate we can't retrieve the data of a vehicle
for payment related methodologies. As RTO(Regional Transport Office) doesn't allow the
general user to collected the confidential information.
1.4 Summary
Through our system we are trying to reduce any engagement related to corruption like bribery
thereby strictly following the traffic rules. Initially our system will detect the QR code along
with number plate of those vehicle who will try to break traffic rules. After that vehicle owner
will receive a payment request asking for the fine. If the owner doesn’t pay the stated amount
then his/her every vehicle document will be declared null and void.

CHAPTER 2
LITERATURE SURVEY
1. Paper Title - Real-Time Illegal Parking Detection System Based on Deep Learning
Authors - Xuemei Xie, Chenye Wang, Shu Chen, Guangming Shi, Zhifu Zhao
Illegal parking has become a growing problem in urban areas, leading to traffic congestion,
safety hazards, and inconvenience for other road users. Current methods for detecting
illegally parked vehicles, such as background segmentation, have limitations in robustness
and sensitivity to the environment. In this paper, the authors propose a novel deep learning-
based system for detecting illegally parked vehicles, leveraging the Single Shot MultiBox
Detector (SSD) algorithm.
The proposed system first locates and classifies illegally parked vehicles using the SSD
algorithm and then optimizes it by adjusting the aspect ratio of default boxes. Next, the
system adopts a tracking and movement analysis approach to identify illegal parking in the
region of interest (ROI). The experiments conducted on real-world datasets demonstrate the
effectiveness of the proposed system, achieving a high accuracy of 99% and real-time
detection at 25 frames per second, even in complex environments.
This study contributes to improving traffic management and safety by providing a more
robust and accurate method for detecting illegal parking. Future work could explore the
integration of this system with other traffic management systems and the potential for
automated enforcement measures to address the problem of illegal parking more effectively.
Merits
The system achieves a high accuracy rate of 99%, which indicates that it can reliably detect
illegally parked vehicles.The system can operate at 25 frames per second, which means that it
can process a large number of video frames in real-time, making it practical for use in urban
environments..
Limitations
The proposed system focuses on detecting illegally parked vehicles and does not address
other issues related to parking management, such as parking time limits or fee collection.

2. Paper Title – VEHICLE NUMBER PLATE DETECTION USING PYTHON AND

OPEN CV
Author - A S Mohammed Shariff, Raghav Bhatia, Raghwendra Kumar and Sarthak
Jha
At the time of high economic growth period of Japan in the 1960s, the supermarkets were
selling a wide range of goods that are foodstuff, clothing, household accessories etc. At the
cash counters, the price of these goods had to enter manually in the cash registers.
Consequently, many cashiers faced the wrist problems and carpal tunnel syndrome. So the
cashiers wanted to find some way to reduce the burden of this manual typing. The invention
of barcodes was the first step towards the solution of this problem and the use of barcodes
gave a relief to the cashiers. When the codes printed on the product were scanned by optical
sensor then the price and other basic information of that product was automatically displayed
on the cash register. As the use of barcodes spread, one of the prominent limitations came in
the way that a barcode can only store 20 alphanumeric characters of information. But the
users were demanding more storage capacity in the barcodes and then after many years of
research, Denso Wave Corporation developed a new type of code which has greater storage
space than earlier barcodes and has fast readability also. Hence the code was named as
“Quick Response code”.
Merits
This system we can easily detect and captures the number plate of a fast moving vehicle. The
system easily convert the picture of the number plate which it capture into a string which can
be processed further.
Limitations
Evening after capturing the number plate we cannot get any information about the vehicle
owner as it is a privileged information. Capturing of number plate will be difficult if it is not
particularly visible to the camera or if somehow the vehicle number is hidden.

3. Paper Title: ENCRYPTED QR CODE GENERATOR

Authors - Vinit Kumar and Manish Kumar Dubey
With the rocketing Indian population, the number of unlicensed vehicles has witnessed a
meteoric rise only aggravating the ever present and persistent traffic problem in the country.
This led to a rise in traffic jams and a significant increase in crimes. So, it becomes important
to perforce introduce a system to quickly hand out fines. In commercial establishments to
permit only authorized vehicles demands substantial number of resources in both financial as
well as in terms of time. So, what we need is a robust and efficient system which will capture
the image of license plate and will use it to extract information from it.Automatic number
plate recognition is a method which is used to extract the characters of license plate numbers
from an image. Image processing is used to extract implement this method. The main steps
involved in image processing are:. Scaling the input image and if the image is in RGB
formatit is converted into Grayscale. Image restoration and enhancement. Prepossessing the
image which involves many methods like morphological processing, image compression.
Segmentation procedure which involves the image partition into its constituent parts. Last
step involves object detection and recognition. This project is also based on above steps.
Various other methods like canny edge detection and bilateral filter are also used in our
project. The image is passed through various methods of image processing and at last
tesseract engine is usedto read the text obtained from the final image.
Merits
The QR code possess high storage capacity means it can store 7089 characters of information.
If the part of code symbol is damaged or dirty, data can be recovered. The error detecting
procedure can focus on the region of correct information.
Limitations
The process may be time consuming because after receiving the code, we need to decrypt it..

4. Paper Title: Convolutional Neural Network-based Automatic Extraction and Fine

Generation
Authors - Y Mohana Roopa ,Sri Harshini Popuri ,Gottam Gowtam sai Sankar,Tejesh
Chandra Kuppili.
It is widely noticeable that from past years the usage of vehicles increased tremendously,
there are a greater number of vehicles than people on road. In the present world, people are
too busy and everyone is in a great hurry. To reach destinations on time, people are breaking
traffic rules. As there so no proper road lane system used in many countries. Most of the
accidents are caused by heavy vehicles when they are around small two-wheelers. Compared
to other vehicles two wheelers are more liable to accidents. The major reason for fatal
accidents on two wheelers is because people are two negligent to use helmets. During the past
decade, the fatality caused by the two wheelers has increased enormously. almost every
country has a separate police force related to traffic. But it’s mostly a humanized system
which is inefficient as the traffic is huge. This can be reduced with a few changes in the
system. Likeusing technology to increase the surveillance and provide a proper system to
manage the surveillance with low cost. This new system will increase carefulness among
people who don’t feel like being captured on CCTV. By this , the number of people using
helmet might increase..
Merits
The major advantage of this method is that it can help in reducing the accident of the two
wheeler vehicles.This will increase in the number of people using helmets those who dot
want to get capture in cctv camera.
Limitations
The disadvantage of this work is that possibly utilize geometric highlights to check if any security
head protector exists in the set.This will increase in the number of people using helmets those who
dot want to get capture in cctv camera.

2.2 Motivation
The motivation behind research on traffic management automation is to improve traffic safety,
reduce congestion, and increase overall efficiency of transportation systems. Traffic
management is a complex task that involves monitoring, controlling, and optimizing traffic
flow. Traditional methods of traffic management, such as human-operated traffic signals and
road signs, are limited in their ability to handle the growing volume of vehicles on the road.
Automation technology, such as artificial intelligence, computer vision, and machine learning,
offers a promising solution to the challenges of traffic management. By analyzing real-time
traffic data and predicting traffic patterns, automated systems can adjust traffic flow, reduce
delays, and prevent accidents. Furthermore, automation can also reduce human error, increase
response time, and improve communication among different transportation stakeholders.
Research in traffic management automation is crucial to advancing the development and

implementation of intelligent transportation systems. By leveraging technology to optimize
traffic flow, we can reduce traffic-related deaths and injuries, increase efficiency, and
improve the overall quality of life in urban areas..
2.3 Summary
The above chapter comprises of the study and reference of various papers related to the project
topic, increasing the overall understanding of the domain and the motivation behind this project
work. The next chapter explains the problem statement, and the system proposed in this project.

CHAPTER 3
PROBLEM STATEMENT AND PROPOSED SYSTEM
3.1 Problem Statement

Traffic amercement is a major problem in many cities around the world. The main problem
lies in the inefficient and time-consuming process of manually issuing traffic tickets to
offenders. This process requires significant manpower and resources and is often prone to
errors and corruption. The backlog of unpaid tickets often leads to a large amount of lost
revenue for cities, impacting their ability to fund necessary infrastructure projects.
Furthermore, the lack of an efficient traffic amercement system can lead to an increase in
traffic violations and unsafe driving practices. Therefore, there is a need for a more automated
and efficient traffic amercement system that utilizes technology to accurately identify traffic
violations and issue fines in a timely and efficient manner.
3.2 Proposed System

The proposed system for traffic amercement automation involves the use of computer vision
and machine learning techniques to automate the process of issuing traffic violations. The
system uses cameras installed at various locations to capture images and videos of traffic
violators. These images and videos are then processed using computer vision algorithms to
detect and recognize license plates of the vehicles and identify the type of violation.
The system also includes a machine learning model trained on a dataset of traffic violations to
classify the type of violation and determine the appropriate fine. The system can also
automatically generate and send a ticket to the violator's registered address through email or
SMS.
The proposed system has several advantages over the traditional manual system of traffic
violation detection and issuing fines. It reduces human error, ensures fair and consistent
enforcement of traffic rules, and saves time and resources. The system can also provide real-
time traffic violation data to authorities, which can be used to improve traffic management
and reduce congestion.

Overall, the proposed system for traffic amercement automation is a promising solution to
improve the efficiency and effectiveness of traffic violation detection and enforcement.
tumors.
3.3 Hardware Requirements

• Processor: Intel Core i5
• RAM Size: 8 GB
• Keyboard: Standard QWERTY serial
• Mouse: Standard serial or PS/2 mouse
• Hard disk: 500 GB
3.4 Software Requirements

• Operating System: Windows 10
• Platform: Visual Studio Editor
• Languages Used: Python
• Domain: Machine Learning, web development
3.5 Applications
The application of traffic amercement system is far-reaching and has significant impact on
road safety, traffic flow management, and law enforcement. With the implementation of
automated systems, traffic law violators can be identified and fined without the need for
physical intervention by traffic police, reducing the risk of accidents and improving road
safety.
Moreover, traffic amercement systems also serve as a deterrent to traffic law violators,
thereby reducing the frequency of traffic violations and improving compliance with traffic
regulations. The data collected by these systems can also be used for traffic flow analysis and
planning, enabling the authorities to make informed decisions regarding traffic management.
In addition to the direct benefits of the system, the automation of traffic amercement also
leads to increased efficiency in the collection of fines and reduced workload for traffic police.
This enables traffic police to focus on more critical tasks such as ensuring public safety and

reducing crime on the roads.
Overall, the application of traffic amercement systems provides a comprehensive solution for
traffic law enforcement and management, and has the potential to greatly improve road safety
and efficiency, while reducing the workload of traffic police.
3.6 Summary
A brief description of the problem statement and the proposed system followed by the
applications of the proposed system are discussed in the above chapter. The next chapter
discusses the high level design of the proposed system. It helps us understand the overall
working.

CHAPTER 4
High Level Design
4.1 System Architecture
The system architecture for traffic amercement automation with fine generation involves
several components working together seamlessly. The architecture comprises three main
modules: the detection module, the back-end module, and the fine generation module.
The detection module uses computer vision techniques such as object detection to detect
traffic violations. It takes input from cameras installed at strategic points. The detection
module then sends the detected violations to the back-end module.
The back-end module takes the users data or the data which that module got from the
detection module and after that it checks whether that vehicle is being already fined or
not.Then it send further the users data to fine generation module.
The fine generation module takes Users data and generates the appropriate fine amount based
on the traffic rules violated. It applies pre-defined rules and regulations to calculate the fine
amount, which can vary based on the severity of the violation. The fine generation module
then sends the fine amount to the payment gateway for further processing.
The architecture also includes a database component that stores all the details of detected
violations, including vehicle number, time of violation, type of violation, and fine amount.
The database can be used for generating reports and statistics on traffic violations.
To ensure the smooth functioning of the system, a central server coordinates the
communication between the various modules. It ensures that data is transmitted efficiently
between modules and that the system is always running optimally.
Overall, the system architecture for traffic amercement automation with fine generation is a
sophisticated and comprehensive solution to tackle the growing problem of traffic violations.
It leverages state-of-the-art technologies such as deep learning, object tracking, and databases

to provide accurate and efficient detection, tracking, and fine generation capabilities.
4.2 System Design
The system design for traffic amercement automation involves the integration of various
components such as cameras, image processing algorithms, and a central database. The
cameras capture images of vehicles violating traffic rules, which are then processed using
computer vision algorithms to extract relevant information. This information is then stored in
a centralized database for further analysis and processing. The system also includes a user
interface for accessing the database and generating automatic fines. The design ensures that
the system is scalable, robust, and accurate in detecting violations.After the violation data is
sned to backend for fine generation and after that that fine is send to users mail. It aims to
minimize human intervention and reduce errors in the traffic amercement process, leading to
better traffic management and safer roads.
Figure 4.1: Level 0 Data Flow Diagram
As shown in Level 0 DFD, the first stage is Data Extraction in whichdata is extracted from Qr
code using ml model and then categorized accordingly, then we have API endpoint and finally
the Backend part taking into consideration user data and fine generation and feedback
mechanism.

Figure 4.2: Level 1 Data Flow Diagram
As shown in the Level 1 DFD, the first process is to take the qrcode encrypted data and send
to backend via api endpoint when any rule is violated. After that fine is generated based upon
the rule violated and send to user’s mail. After payment a feedback form is send to user’s mail
to dhare thoer experience.
4.3 Modules identified.
This project is mainly divided into 4 modules :

 Qrcode Generation
 Vehicle Detection
 Users Information Extraction
 Fine Genereation
 FeedBack Mechanism

1. Qrcode Generation :
QR code generation is the process of creating QR codes, which are two-dimensional

barcodes that can store large amounts of informations. The process typically involves
encoding the information into a QR code format and generating the code using
software or online tools.
2. Vehicle Detection:
Vehicle detection using Haar Cascade Classifier is a computer vision technique that
uses a pre-trained model to detect the presence of vehicles in images. The technique
involves training a classifier to detect the presence of vehicles based on features such
as shape, size, and texture.
3. Users Information Extraction:
Fernet encryption is a symmetric encryption algorithm that generates a secure

message token. It uses a secret key to encrypt data, which can then be decrypted using
the same key. Users data is encrypted using this algorithm and stored into QR.
4. Fine Generation:
The fine generation module is a part of a system that automatically generates fines for
traffic violations. It processes the data collected from the violation detection module
and generates a fine amount based on the severity of the violation. Once the fine is
generated, it is sent to the violator through a notification system(via email).
5. FeedBack Mechanism:
The purpose of a feedback mechanism is to help organizations or individuals identify

areas for improvement and take corrective action to achieve better results.In this a
feedback form is send to users email for their experience with our system.

4.4 Summary
The traffic amercement automation system's high-level design includes vehicle detection
through a camera, fine generation, and an online payment system. It also incorporates a
feedback mechanism for users to contest false fines..

CHAPTER 5
LOW LEVEL DESIGN
5.1 Algorithms Used
In this project, two main algorithms are utilized to achieve the desired outcome - Haar Cascade
Classifier and Fernet algorithm. The Haar cascade algorithm is a machine learning-based
approach used for object detection. It uses a set of positive and negative images to train a
classifier and generate a cascade of classifiers, which are used to detect objects of interest in a
new image. On the other hand, Fernet is a symmetric encryption algorithm that provides
confidentiality by encrypting data using a secret key. It is a part of the cryptography library in
Python, and its implementation uses AES in CBC mode with a 128-bit key for encryption and
HMAC using SHA256 for authentication. Fernet encryption provides integrity,confidentiality,
and authentication. It is a lightweight and easy-to-use encryption algorithm, and it is widely
used in securing data in transit and storing sensitive information..
5.1.1 Haar Cascade

Haar Cascade Classifier is an object detection algorithm used to detect the presence of an
object in an image. In this case, the object is a car. The algorithm uses an XML file which
contains the features of the object it is trained to detect.
1. The XML file containing the Haar features for car detection is loaded.
2. A connection to the default camera is opened, and frames are continuously captured.
3. Each frame is converted to grayscale, and the algorithm detects vehicles using the Haar
Cascade Classifier detectMultiScale function.
4. The detectMultiScale function applies the Haar feature classifiers at different scales to the
image, looking for the presence of the object.
5. If a car is detected, a bounding box is drawn around it in green.
6. The frames with the bounding boxes are displayed in real-time on the screen.
7. If the 'q' key is pressed, the loop breaks, and the camera connection is released, and all
windows are closed.

8. The algorithm can be used for various applications, such as surveillance, parking
management, and traffic monitoring..
5.1.2 Fernet Algorithm :
Fernet algorithm is a symmetric encryption algorithm used to protect data integrity and
confidentiality. It is based on the Advanced Encryption Standard (AES) and uses a shared
secret key to encrypt and decrypt data.
1. The Fernet algorithm is a symmetric encryption algorithm that uses the AES encryption
algorithm in CBC mode with a 128-bit key for encryption and decryption.
2. The Fernet class in the cryptography library provides an implementation of the Fernet
algorithm.
3. To create a Fernet instance, you need to generate a key using the Fernet.generate_key()
method. This key is a base64-encoded 32-byte (256-bit) key that will be used for
encryption and decryption.
4. To encrypt a message using Fernet, you first need to create a Fernet instance with the key
that you generated, and then call the encrypt() method on the instance, passing in the
message as a bytes object.
5. The encrypt() method first generates a random 128-bit initialization vector (IV) and
encrypts it along with the message using AES in CBC mode with the key you provided.
6. The encrypted message is then returned as a bytes object that includes the base64-encoded
IV and the encrypted message.
7. To decrypt a message that was encrypted with Fernet, you first create a Fernet instance
with the same key that was used to encrypt the message.
8. You then call the decrypt() method on the Fernet instance, passing in the encrypted
message as a bytes object.
9. The decrypt() method first extracts the base64-encoded IV and encrypted message from the
input bytes object.
10. It then decrypts the encrypted message using the IV and key you provided, using AES in
CBC mode.
11. The decrypted message is then returned as a bytes object.
12. The Fernet algorithm provides a simple, secure way to encrypt and decrypt messages
using a shared secret key. It is widely used in applications that require secure
communication over insecure channels, such as web applications, messaging systems, and
file storage systems.

5.2 Methodology
The aim of traffic amercement automation is to develop a system that automates the process
of issuing traffic violation tickets, thus reducing the time and effort required by law
enforcement officials. The system is designed to detect traffic violations such as over
speeding, running a red light, and illegal parking using computer vision techniques. Once a
violation is detected, the system generates a ticket that includes the details of the violation
and the fine amount. The ticket is sent to the violator through email, along with an option to
pay the fine online. The ultimate goal of this system is to ensure better traffic management,
reduce road accidents caused by traffic violations, and improve the overall safety and
efficiency of the transportation system.
Methodology:
1. Firstly, Qr code is generated with users data encpryted inside it with the help of Fernet
Encryption algorithm.
2. When any traffic violation is done by any vehicle their Qr code is scanned and users data is
extracted.
3. That user’s data is send to backend via an api endpoint.
4. From their fine is genertaed based on the violation done them.
5. Fine is send to User’s Email , that can be paid online.
6. Later after fine is paid they will get the feedback mail .
Figure 5.1: Methodology flow diagram

CHAPTER 6
IMPLEMENTATION
6.1 Procedure Involved In Execution of Project
The procedure involved in the execution of traffic amercement automation typically includes
the following steps:
1. Capturing Images: First, a camera system is used to capture images of vehicles that violate
traffic rules.
2. Vehicle Detection: Next, the images are processed using the Haar Cascade classifier to
detect vehicles in the images.
3. QR Code detection: The process of QR code detection involves locating the QR code in the
image or video frame, extracting the code's pattern or layout, and decoding the
information stored in the code.
4. Fine Calculation: The fines for the traffic violations are then calculated based on the type of
violation.
5. Payment and Feedback: The system then generates a fine and sends it to the vehicle owner.
The system also provides a feedback mechanism to allow vehicle owners to dispute or
pay the fine.
6. Data Analysis: Finally, the data collected from the system can be analyzed to identify
patterns and trends in traffic violations, which can be used to improve traffic safety
measures..
6.2 Module1. Qr Code Generation and Encryption
 QR code generation is the process of creating a two-dimensional barcode that can be scanned
using a QR code reader or smartphone app. The QR code typically contains some form of
data or information, such as a URL, contact information, or other data.
 In this module we generate the Qr code and stores the user info in the Qr after encrypting the
info.
 Here we use fernet algorithm to first encrypt the data using a symmetric key.
 Once the encryption has been done then we pass to encrypted data to Qr generation
module ,where it generate the Qr code.
 After the we create the image form of the Qr code and store it in our current directory

6.3 Module 2. Qr Code Decryption
 After the encrypted Qr code is generated,our current module then decrypt the
contents of the Qr code.
 Then it capture the content of the Qr code through camera and decode the content
and store it in a python dictionary.
 The information present in our python dictionary is sent to our backend server
through an api enpoint exposed in our module.
 After the Qr code has been scanned our camera will automatically closes and futher
processing will happen in our backend server.
6.4 Module 3. Fine Generation
 Send the User’s data to the backend: Use an HTTP request library, such as requests
in Python, to send the phone number to the backend API endpoint. You can send the
phone number as a parameter in the request URL or in the request body.
 Call the payment API: Once the backend receives the phone number, it can then call
the appropriate payment API, passing in the phone number as a parameter. This will
initiate the payment request.
 Handle the payment response: Once the payment API returns a response, the backend
can handle it accordingly. This may involve updating a database or sending a
notification to the user.
Fig:6.1 Fine request is send to phone number of user’s

Fig:6.2 Fine request is send to users Email
 Keeping the record of Payment: All the details of rule violators will be displayed in
tabular format on the E-challan website with help of MUI library and ReactJs.The
system also keeps track of the payment link status and user data in the database,
ensuring efficient and accurate record keeping.
Fig:6.3 E-challan website
6.5 Module 4. FeedBack Mechanism

 Collect payment details: Collect the payment details from the customer, such as the
amount, payment method, and billing information.
 Process the payment: Use a payment processing algorithm to securely process the
payment and authorize the transaction.
 Provide feedback to the customer: Provide feedback to the customer on the status of
the payment, such as whether it was successful or declined.
 Record the transaction: Record the transaction in a database or accounting system for
future reference.
 Handle disputes: Handle any disputes or issues that may arise with the payment, such
as charge backs or fraud.
.
CHAPTER 7
SYSTEM TESTING
Testing is an important phase in the development life cycle of the product. This is the phase,
where the remaining errors, if any, from all the phases are detected. Hence testing performs a
very critical role for quality assurance and ensuring the reliability of the software.
During the testing, the program to be tested was executed with a set of test cases and the output
of the program for the test cases was evaluated to determine whether the program was
performing as expected. Errors were found and corrected by using the below stated testing
steps and correction was recorded for future references. Thus, a series of testing was performed
on the system, before it was ready for implementation.
It is the process used to help identify the correctness, completeness, security, and quality of
developed computer software. Testing is a process of technical investigation, performed on
behalf of stake holders, i.e. intended to reveal the quality-related information about the product
with respect to context in which it is intended to operate. This includes, but is not limited to,
the process of executing a program or application with the intent of finding errors.
The quality is not an absolute; it is value to some person. With that in mind, testing can never
completely establish the correctness of arbitrary computer software; Testing furnishes a
‘criticism’ or comparison that compares the state and behavior of the product against
specification. An important point is that software testing should be distinguished from the
separate discipline of Software Quality Assurance (SQA), which encompasses all business
process areas, not just testing.
There are many approaches to software testing, but effective testing of complex products is
essentially a process of investigation not merely a matter of creating and following routine
procedure.
Although most of the intellectual processes of testing are nearly identical to that of review or
inspection, the word testing is connoted to mean the dynamic analysis of the product-putting
the product through its paces. Some of the common quality attributes include capability,
reliability, efficiency, portability, maintainability, compatibility and usability.
A good test is sometimes described as one, which reveals an error; however, more recent
thinking suggest that a good test is one which reveals information of interest to someone who
matters within the project community.

System testing of software or hardware is testing conducted on a complete, integrated system

to evaluate the system's compliance with its specified requirements. The purpose of testing is
to discover errors. Testing is the process of trying to discover every conceivable fault or
weakness in a work product. It provides a way to check the functionality of components, sub-
assemblies, assemblies and/or a finished product. It is the process of exercising software with
the intent of ensuring that the software system meets its requirements and user expectations,
and does not fail in an unacceptable manner. There are various types of testing. Each test type
addresses a specific testing requirement.
7.1 Types of Testing
7.1.1 Unit Testing

The primary goal of unit testing is to take the smallest piece of testable software in the
application, isolate it from the remainder of the code, and determine whether it behaves
exactly as you expect. Intuitively, one can view a unit as the smallest testable part of an
application. Each unit is tested separately before integrating them into modules to test the
interfaces between modules. Unit testing involves the design of test cases, which validate
that the internal program logic is functioning properly and that the program inputs produce
valid outputs. All decision branches and internal code flow should be validated. It is process
of testing the individual software units of the application. It is done after the completion of
an individual unit and before integration.
With regard to the proposed system, each sensor should work properly to get the proper
sensor data values..
7.1.2 Integration Testing
Integration testing is the phase in software testing in which individual software modules are
combined and tested as a group. It occurs after unit testing and before validation testing.
Integration testing takes as its input modules that have been unit tested, groups them in larger
aggregates, applies tests defined in an integration test plan to those aggregates, and delivers as
its output the integrated system ready for system testing. Integration testing is designed to test
integrated software components to determine if they actually run as one program. Integration
testing demonstrates that although the components were individually satisfied by successful

unit testing, the combination of components is correct and consistent. Integration testing is
specifically aimed at exposing the problems that arise from the combination of components. In
the proposed system all the components should work as a whole program to achieve the goals
for which it is implemented. For example, all the sensors should work together when connected
along with each other which determines whether the components are working properly together
to get the result.
7.1.3 Functional Testing

Functional testing is a quality assurance (QA) process and a type of black box testing that bases
its test cases on the specifications of the software component under test. Functions are tested
by feeding them input and examining the output, and internal program structure is rarely
considered (not like in white-box testing). Functional Testing usually describes what the
system does. Functional tests provide systematic demonstrations that functions tested are
available as specified by the business and technical requirements, system documentation and
user manuals.
Organization and preparation of functional tests is focused on requirements, key functions, or
special test cases. In addition, systematic coverage pertaining to identify business process
flows, data fields, predefined processes and successive processes must be considered for
testing. Before functional testing is complete, additional tests are identified and the effective
value of current tests is determined.
7.1.4 System Testing

to evaluate the system's compliance with its specified requirements. System testing falls within
the scope of black-box testing, and as such, should require no knowledge of the inner design
of the code or logic.
As a rule, system testing takes, as its input, all of the "integrated" software components that
have passed integration testing and also the software system itself integrated with any
applicable hardware system(s). The purpose of integration testing is to detect any
inconsistencies between the software units that are integrated together (called assemblages) or
between any of the assemblages and the hardware. System testing is a more limited type of
testing; it seeks to detect defects both within the "inter-assemblages" and also within the system
as a whole.
to evaluate the system's compliance with its specified requirements. It tests a configuration to
ensure known and predictable results. System testing is based on process descriptions and
flows, emphasizing pre-driven process links and integration points.
7.1.5 Black Box Testing
Black-box testing is a method of software testing that examines the functionality of an

application (e.g. what the software does) without peering into its internal structures or workings
(see white-box testing). This method of test can be applied to virtually every level of software
testing: unit, integration, system and acceptance. The tester is oblivious to the system
architecture and does not have access to the source code. Typically, when performing a black
box test, a tester will interact with the system's user interface by providing inputs and
examining outputs without knowing how and where the inputs are worked upon. It’s most
widely used because its well suited and efficient for large code segments, code access is not
required, clearly separates user's perspective from the developer's perspective through visibly
defined roles, large numbers of moderately skilled testers can test the application with no
knowledge of implementation, programming language or operating systems
7.1.6 White Box Testing
White-box testing (also known as clear box testing, glass box testing, transparent box testing,
and structural testing) is a method of testing software that tests internal structures or workings
of an application, as opposed to its functionality (i.e. black-box testing). In white-box testing
an internal perspective of the system, as well as programming skills, are used to design test
cases. The tester chooses inputs to exercise paths through the code and determine the
appropriate outputs. White box testing is the detailed investigation of internal logic and
structure of the code. White box testing is also called glass testing or open box testing. In order
to perform white box testing on an application, the tester needs to possess knowledge of the
internal working of the code. The tester needs to have a look inside the source code and find
out which unit/chunk of the code is behaving inappropriately.

7.1.7 Test Cases
Fig:7.1 Unpaid or created fine
Fig:7.2 Paid fine
7.2 Summary
The above section consists of the various kinds of testing that are carried out for checking and
ensuring proper working of the system. Different images are given as input and checked if the
fine is paid or not by a user. The next chapter concludes the project.

CHAPTER 8
CONCLUSION
In conclusion, the implementation of traffic amercement automation can have a significant

impact on the management of traffic violations and improving road safety. By integrating
various technologies such as Haar Cascade classifiers, QR code scanning, and Fernet
encryption, the system can accurately detect and fine traffic offenders while ensuring the
security of sensitive data.
The proposed system can also provide real-time monitoring of traffic violations, allowing for
prompt action to be taken against offenders. The feedback mechanism enables the system to
continuously improve its accuracy, ensuring that false positives are minimized.
Furthermore, the use of QR code scanning can help reduce the administrative burden and
improve the efficiency of the process, as the fines can be automatically generated and linked
to the offender's record.
Overall, the implementation of traffic amercement automation can bring several benefits,
such as reducing road accidents and fatalities, improving traffic flow, and reducing the
workload of traffic police officers. As technology continues to advance, it is imperative that
governments and organizations embrace these innovations to improve public safety and
ensure the smooth functioning of transportation systems..

CHAPTER 9
FUTURE SCOPE
The future scope of traffic amercement automation is promising as it can further be improved
with advanced technology. One potential area for improvement is the integration of artificial
intelligence and machine learning algorithms to enhance the accuracy of traffic violation
detection. These technologies can help to better analyze and classify traffic violations based
on their severity, thereby improving the overall effectiveness of the system.
Furthermore, the integration of blockchain technology can enhance the transparency and
security of traffic violation records, ensuring that they are tamper-proof and easily accessible
by relevant authorities. The use of big data analytics can also provide valuable insights into
traffic patterns and behaviors, which can be used to improve traffic management and
infrastructure planning.
Overall, the future scope of traffic amercement automation is vast, with numerous
opportunities for improvement and expansion. As technology continues to evolve, it is likely
that traffic management systems will become more advanced and efficient, leading to safer
and more organized roads for all.

CHAPTER 10
SNAPSHOTS
Figure 10.1 : QR code generation
Fig:10.2 Unpaid or created fine

 The user interface includes a file uploader widget, a prediction button, and an output text box.
When the user uploads an image and clicks the "Predict" button, the model predicts the type of
tumor in the uploaded image and displays the prediction in the output text box. The user
interface is created using the ipywidgets library in Python. The code also includes styling for
the widgets and a logo image.
Figure 10.3 : E-challan
 The output page includes a file uploader, prediction button, and an output text box. When the
user uploads an image and clicks the "Predict" button, the model predicts the type of tumor in
the uploaded image and displays the prediction in the output text box.

REFERENCES
[1] Xuemei Xie, Chenye Wang, Shu Chen, Guangming Shi, Zhifu Zhao of Xidian
University. Real-Time Illegal Parking Detection System Based on Deep Learning in 2017
International Conference.
[2] Y Mohona Roopa & Sri Harishini Popuri Convulation Neural Network-based automatic
extraction and fine generation. In 2020 2nd International Conference on Inventive Research in
Computing Application.
[3] S. Hariswetha, S. Indira A.& S. Latha., QR based Automatic Penalty Charging for
Violation of Traffic Rules and spark streaming. In 2020 International Journal of Engineering
Research & Technology (IJERT).
[4] Vinit Kumar and Manish Kumar Dubey.Encrypted QR code generator. In 2019
International general of recent scientific reaserch.
[5] Lina Hua, JIanxing zhu and Pradeep Kumar Singh. Qr image code recognition system
based on AI algorithm. In 2021 Journal of Intelligent System.

International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 07 Issue: 01 | January - 2023 Impact Factor: 7.185 ISSN: 2582-3930
DESIGN OF TRAFFIC AMERCEMENT AUTOMATION USING

COMPUTER VISION
Chirag Suthar1, Chirantan Banerjee2, Gaurav Mourya3, Ishan Makharia4, .Nagraj M.Lutimath5
1
Student, 2Student, 3Student, 4Student, 5Associate Professor
1
Department of Computer Science and Engineering
1
Dayananda Sagar Academy of Technology and Management, Bengaluru, Karnataka, India
Abstract: Traffic rule violations, such as speeding, not Traffic-related fatalities, particularly those involving two-
wearing a proper protective helmet, and running red lights, wheelers, have risen significantly in the past decade. Most
are a major contributor to the high number of road countries have dedicated traffic police forces, but these
accidents in India. According to National Crime Records systems are often overwhelmed by the sheer volume of
Bureau data, over 1.5 lakh people lose their lives in road traffic and can be inefficient in preventing accidents.
accidents across the country every year, an average of 426
daily or 18 every hour. To address this issue, we propose Developing countries like India is majorly equipped
an automated system for collecting traffic fines using with humanized traffic system. Such system can sometimes
machine learning techniques. The system utilizes object be a blessing and sometime it can be a curse. Here blessing
detection to identify vehicles that have violated traffic rules, comes with the fact that such system is less prone to human
using video surveillance cameras. It then extracts the error but not totally. Curse of such system is one of the main
number plate and an encrypted QR code of the vehicle, reason it should be avoided. Curse involve any
which can be used to retrieve the vehicle owner's phone miscommunication, misdirection and our major topic of
number and other details. These details are stored on a
discussion Bribery. People in such system can easily get
server, and the system sends three reminders for payment
within a fixed time window. If the fine is not paid, the away with their crime which involve over-speeding, not
vehicle's registration with the RTO (Regional Transport wearing helmet etc by simply bribing the traffic police. Such
Office) is automatically cancelled and it is no longer system doesn’t bring any justice to the society as such action
allowed on the road. This system aims to improve the only increases the traffic violation rate.
efficiency and accuracy of the traffic fine collection Computer equipped traffic system can easily out
process, while also reducing errors and the possibility of
perform a humanized system. Here everything can be
bribery. The use of machine learning techniques and video
automated by simply eliminating the need of any middle-
surveillance cameras allows for continuous monitoring of
traffic, reducing the need for large numbers of traffic men. Such system can efficiently implement the traffic rules.
police to manually monitor the roads. The extraction of It can check whether someone is violating the traffic rules of
the QR code and vehicle owner's details also makes it not , if someone is violating then a proper fine is imposed on
easier to track and verify transactions, ensuring that the him/her .
correct fine is being paid by the right individual. The
automatic cancellation of the vehicle's registration if the
fine is not paid serves as a deterrent for individuals II. LITERATURE SURVEY
attempting to evade paying their fines. In summary, our In [1] A S Mohammed research they simply describe the
proposed system utilizes advanced technology to automate process here involves several steps to extract the license
the traffic fine collection process, improving efficiency and plate number from an input image using image processing
accuracy while also reducing the possibility of bribery and techniques. The first step is to convert the input image from
errors.y. RGB to grayscale format. The bilateral filter is then applied
I. INTRODUCTION to remove background noise while preserving edges. Canny
edge detection is used to detect the edges of the license plate.
There has been a significant increase in the number of
All contours in the processed image are detected, and the top
vehicles on the roads in recent years, with many places
30 contours are selected for further processing. These
experiencing more vehicles than people. This increase in
contours are checked for rectangular shape, and if found to
traffic has led to people breaking traffic rules in an effort to be a rectangle, they are considered to be the license plate and
reach their destinations on time. In some countries, the lack the image is cropped accordingly. The cropped image is then
of proper road lane systems can also contribute to accidents, passed through a text recognition tool such as Tesseract ,
particularly when heavy vehicles are near smaller two- which extracts the string of the license plate number and
wheelers. Two-wheelers are particularly prone to accidents displays it on the terminal. This process allows for the
due to their vulnerability on the roads, and the lack of helmet accurate and efficient extraction of license plate numbers
usage is a major contributor to fatalities in this category.
© 2023, IJSREM | www.ijsrem.com | Page 1

from images, enabling the automation of traffic rule violation details of the violation and a payment request, is then sent
detection and fine issuance. to the registered phone number via SMS or UPI app. The
fine is automatically collected from the owner's bank
In a research study conducted by Y Mohona Roopa [2], the account, and the information is updated online for future
primary goal was to fully automate the process of detecting reference.
helmet usage and issuing fines. Previously, traffic
This android-based system provides a convenient and
surveillance was largely semi-automated or completely
efficient way to enforce traffic rules and collect fines. By
reliant on human input. However, the new system being
automating the process, it helps to reduce the workload of
proposed aims to detect vehicles without helmets using
traffic police and promotes compliance with traffic
CCTV footage and extract the license plate. The number is
regulations. It also helps to reduce the number of accidents
then searched in a database to obtain the registered phone
and fatalities on the roads by encouraging responsible
number. The image of the footage, along with details of the
driving behavior.
fine and payment options, is mailed and sent via message to
the registered phone number. The information is also
updated online for future reference. This automated process In [4] The program for generating QR codes is
allows for efficient and accurate enforcement of helmet implemented using the JavaScript language and run on a
usage regulations and promotes compliance with traffic node.js server. The generated QR code is displayed on a
rules.. web browser for the user to view and scan. In order to add
Here in our proposed system we can use the model to an additional layer of security, the QR code is encrypted
detect the number plate of the vehicle along with the qr code using Elliptical Curve Cryptography (ECC).
attach to it. The Qr code will be initially encrypted, we can To scan or read the encrypted QR code, the user needs to
decrypt it to get the required information and then easily install a QR code scanner app on their smartphone. There
sent the payment details to the corresponding vehicle owner. are many such apps available for free on app stores. Once
the app is installed, the user can start the application and
In [3] the proposed system is an android-based solution for use the smartphone camera to scan the QR code. The app
automating the process of detecting and enforcing traffic will automatically display the content of the QR code once
rule violations. It aims to address the challenges associated it has been scanned.
with manual traffic rule enforcement by utilizing QR code This system provides a convenient and secure way for
technology and online payment methods. users to access information encoded in QR codes. The
In this system, a QR reader is used to scan the QR code encryption ensures that the information cannot be accessed
attached to each vehicle. This code, which is mandatory by unauthorized parties, while the QR code scanner app
for all vehicles, is obtained when the vehicle is purchased allows for easy and quick access to the information by the
or when the driver receives their license from the Road intended user.
Transport Office (RTO). When a violation is detected, the
system extracts the QR code from the footage and uses it
In[5],First, a selection of QR code images were chosen and
to search for the registered phone number in a database.
the mobile phone QR code reading software was used to
An image of the footage, along with details of the violation
try and read the information contained within them. The
and a payment request, is then sent to the registered phone
statistical recognition rate of the QR code images that
number via SMS or UPI app. The fine is automatically
could not be read by the software was then analyzed.
collected from the owner's bank account, and the
information is updated online for future reference. Next, the processed QR code images were read using the
mobile phone QR code reading software, and the statistical
This android-based system provides a convenient and
recognition rate was analyzed. The results showed that the
efficient way to enforce traffic rules and collect fines. By
QR code reading rate after processing by the system
automating the process, it helps to reduce the workload of
algorithm in this article increased by 14%. This
traffic police and promotes compliance with traffic
demonstrates the effectiveness of the algorithm in
regulations. It also helps to reduce the number of accidents
improving the recognition rate of QR code images.
and fatalities on the roads by encouraging responsible
driving behavior.In this system, a QR reader is used to However, it should be noted that the processed images
scan the QR code attached to each vehicle. This code, were not recognized by the existing reading software,
which is mandatory for all vehicles, is obtained when the indicating that further research is needed to explore the
vehicle is purchased or when the driver receives their security and application aspects of the algorithm in other
license from the Road Transport Office (RTO). When a areas. In the future, image enhancement and robustness
violation is detected, the system extracts the QR code from issues may also be analyzed in more detail to further
the footage and uses it to search for the registered phone improve the performance of the QR code reading
number in a database. An image of the footage, along with algorithm.

generalizability. By increasing the number of images used

III. PROPOSED METHODOLOGY for training, it is possible to further improve the model's
accuracy
C. Implementation through opencv
OpenCV is a tool for image processing that can be used to
identify objects. In this system, the OpenCV library is used
to differentiate the trained object (e.g. helmet) from other
objects. This library supports three programming
languages: Java, Python, and C++. By using OpenCV, it is
possible to differentiate and detect helmets usage in
images. The library's various functions and algorithms can
be used to analyze and extract information from images,
enabling the identification of specific objects or patterns.
D. Frames Collection
OpenCV is a tool for image processing that can be used to
identify objects. In this system, the OpenCV library is used
to differentiate the trained object (e.g. helmet) from other
objects. This library supports three programming
Fig1: flow diagram of proposed methodology languages: Java, Python, and C++. By using OpenCV, it is
possible to differentiate and detect helmets usage in
The proposed system aims to fully automate the
images. The library's various functions and algorithms can
process of detecting traffic rule violations and issuing fines
be used to analyze and extract information from images,
using CCTV cameras as shown in fig 1. When a violation
enabling the identification of specific objects or patterns.
is detected, the system extracts the vehicle's license plate
or QR code from the footage and searches for the E. Number Plate Extraction
associated phone number in a database. The system then The process of extracting number plates from images or
sends an image of the footage, along with details of the video frames uses an automatic number plate recognition
violation and a UPI payment request, to the registered (ANPR) module. This module utilizes an API from Plate
phone number. The information is also updated online for Recognizer to analyze the images and identify the number
future reference. The automated process involves several plates. The locator module sends a request for the image or
stages, including the detection of the violation, video frame being processed to the ANPR module, and
identification of the vehicle, and issuance of the fine.. receives a JSON response. The field containing the
A. Violation Recognition vehicle's number plate in the frame is then extracted using
standard Python techniques. The ANPR module is able to
The YOLOv3 object detection framework is used to detect accurately and efficiently identify and extract number
traffic violations such as helmet usage and over speeding. plates from a wide range of images and video frames,
This neural network-based system analyzes images at test making it a useful tool for automated traffic rule violation
time to detect the presence of objects. In the proposed detection systems.
system, videos from traffic surveillance are processed
using YOLOv3. This framework is particularly useful for F. QR Code Extraction
detecting a wide range of objects, including two-wheelers, In the proposed system, the QR code associated with a
and is known for its speed and accuracy. In this system, vehicle is scanned and decrypted to extract the registered
YOLOv3 is trained to identify instances of two-wheeler mobile number. Once the mobile number has beenobtained,
vehicles that are not wearing helmets or are speeding. a fine is generated in the database for the corresponding
B. Training over model traffic rule violation. The number plate and the final frame,
along with details of the fine, are updated on the server. A
To train a custom helmet or identification model, a large message containing a UPI payment request is then sent to
number of images showing helmets usage in various the registered phone number via SMS gateway or UPI app.
situations will be needed. The more images that are used This automated process allows for efficient and convenient
for training, the more precise the model will be. It is likely payment of fines, and helps to ensure that traffic rule
that several thousand images will be required to achieve violations are effectively enforced.
sufficient accuracy. It is important to ensure that the
images used for training represent a diverse range of G. Fine Generation
situations, including different lighting conditions, angles, The proposed system is able to generate fines for traffic
and environments, in order to improve the model's rule violations by detecting the vehicle's number plate or

by scanning the QR code associated with the vehicle.

[4] Vinit Kumar and Manish Kumar Dubey.Encrypted QR code
When a violation is detected, the system extracts the
generator. In 2019 International general of recent scientific
number plate or QR code from the footage and searches reaserch.
for the associated phone number in a database. The fine is
then automatically generated and sent to the registered [5] Lina Hua, JIanxing zhu and Pradeep Kumar Singh. Qr image
phone number along with details of the violation and a code recognition system based on AI algorithm.
payment request. This automated process allows for In 2021 Journal of Intelligent System.
efficient and accurate enforcement of traffic rules, and
helps to promote compliance with traffic regulations. It
also allows for convenient payment of fines through the
use of UPI apps or SMS gateway.
IV Conclusion
Automating the process of collecting traffic fines using
number plate recognition technology and qr code
encryption can have several benefits. One of the main
advantages is that it can make the process more efficient
and convenient for both the authorities and the individuals
who have received the fines. With QR code recognition,
authorities can easily identify vehicles that have been
involved in traffic violations and automatically send a fine
to the owner of the vehicle by extracting data from
Encrypted QR code. This can save time and effort for both
parties, as the individual does not have to physically visit a
payment center to pay their fine.
Another benefit is that automating the fine collection
process can help to reduce errors and fraud. By using
number plate recognition technology, it is easier to track
and verify transactions, ensuring that the correct amount is
being paid by the right person. This can help to prevent
individuals from trying to evade paying their fines or from
being wrongly charged.
Overall, automating the process of collecting traffic fines
using number plate recognition and QR code technology
can improve the efficiency and accuracy of the process,
making it more convenient for both the authorities and
individuals. It can also help to reduce errors and fraud,
ensuring that the correct amount is being paid by the right
person.
Reference:
[1] A S Mohammed Shariff Vehicle number plate detection
using python and open cv. In 2021 Internation Conference on
Advance computing and Inovativetechnology I Engineering.
[2] Y Mohona Roopa & Sri Harishini Popuri Convulation Neural

Network-based automatic extraction and fine generation. In
2020 2nd International Conference on Inventive Research in
Computing Application.
[3] S. Hariswetha, S. Indira A.& S. Latha., QR based

Automatic Penalty Charging for Violation of Traffic
Rules and spark streaming. In 2020 International Journal
of Engineering Research & Technology (IJERT).

INTERNATIONAL JOURNAL OF ADVANCE
RESEARCH AND INNOVATIVE IDEAS IN EDUCATION
The Board of International Journal of Advance Research and Innovative Ideas in Education
is hereby Awarding this Certificate to
GAURAV MOURYA
IJARIIE In Recognition of the Publication of the Paper Entitled

DESIGN OF TRAFFIC AMERCEMENT AUTOMATION USING COMPUTER VISION
Published in E-Journal
Volume-9 Issue-3 2023
Paper Id : 20141
ISSN(O) : 2395-4396 Editor In Chief
www.ijariie.com
ISHAN MAKHARIA

Paper Id : 20141
www.ijariie.com
CHIRAG SUTHAR

Paper Id : 20141
www.ijariie.com
CHIRANTAN BANERJEE

Paper Id : 20141
www.ijariie.com
MRS ARPITHA VASUDEV

Paper Id : 20141
www.ijariie.com
Vol-9 Issue-3 2023 IJARIIE-ISSN(O)-2395-4396
DESIGN OF TRAFFIC AMERCEMENT

AUTOMATION
USING COMPUTER VISION
Mrs Arpitha Vasudev1,Chirag Suthar2, Chirantan Banerjee3, Gaurav
Mourya4, Ishan Makharia5
1Assitant Professor, 2Student, 3Student, 4Student, 5Student
1Department of Computer Science and Engineering
1Dayananda Sagar Academy of Technology and Management, Bengaluru,
Karnataka, India
Abstract
Traffic rule violations, such as speeding, not wearing a proper protective helmet, and running red lights, are
a major contributor to the high number of road accidents in India. According to National Crime Records
Bureau data, over 1.5 lakh people lose their lives in road accidents across the country every year, an
average of 426 daily or 18 every hour. To address this issue, we propose an automated system for
collecting traffic fines using machine learning techniques. The system utilizes object detection to identify
vehicles that have violated traffic rules, using video surveillance cameras. It then extracts the number plate
and an encrypted QR code of the vehicle, which can be used to retrieve the vehicle owner's phone number
and other details. These details are stored on a server, and the system sends three reminders for payment
within a fixed time window. If the fine is not paid, the vehicle's registration with the RTO (Regional
Transport Office) is automatically cancelled and it is no longer allowed on the road. This system aims to
improve the efficiency and accuracy of the traffic fine collection process, while also reducing errors and the
possibility of bribery. The use of machine learning techniques and video surveillance cameras allows for
continuous monitoring of traffic, reducing the need for large numbers of traffic police to manually monitor
the roads. The extraction of the QR code and vehicle owner's details also makes iteasier to track and verify
transactions, ensuring that the correct fine is being paid by the right individual. The automatic cancellation
of the vehicle's registration if the fine is not paid serves as a deterrent for individuals attempting to evade
paying their fines. In summary, our proposed system utilizes advanced technology to automate the traffic fine
collection process, improving efficiency and accuracy while also reducing the possibility of bribery and
errors.y.
I. INTRODUCTION
There has been a significant increase in the number of vehicles on the roads in recent years, with many
places experiencing more vehicles than people. This increase in traffic has led to people breaking traffic
rules in an effort to reach their destinations on time. In some countries, the lack of proper road lane systems
can also contribute to accidents, particularly when heavy vehicles are near smaller two- wheelers. Two-
wheelers are particularly prone to accidents due to their vulnerability on the roads, and the lack of helmet
usage is a major contributor to fatalities in this category. Traffic-related fatalities, particularly those
involving two- wheelers, have risen significantly in the past decade. Most countries have dedicated traffic
police forces, but these systems are often overwhelmed by the sheer volume of traffic and can be inefficient
in preventing
accidents.Developing countries like India is majorly equipped with humanized traffic system. Such system
can sometimes be a blessing and sometime it can be a curse. Here blessing comes with the fact that such
system is less prone to human error but not totally. Curse of such system is one of the main reason it should
be avoided. Curse involve any miscommunication, misdirection and our major topic of discussion Bribery.
20141 ijariie.com 620

People in such system can easily get away with their crime which involve over-speeding, not wearing helmet
etc by simply bribing the traffic police. Such system doesn’t bring any justice to the society as such action
only increases the traffic violation rate.
Computer equipped traffic system can easily out perform a humanized system. Here everything can be
automated by simply eliminating the need of any middlemen. Such system can efficiently implement the
traffic rules. It can check whether someone is violating the traffic rules of not , if someone is violating then a
proper fine is imposed on him/her .
II. PROBLEM DESCRIPTION

Traffic rule violations are a significant problem in India and are responsible for a considerable number
of road accidents, leading to the loss of many lives annually. The existing method of manually monitoring
and collecting traffic fines is vulnerable to inefficiencies, inaccuracies, and corrupt practices. Thus, there is
an urgent need for an automated system that can efficiently and precisely collect traffic fines while
minimizing errors and reducing the possibility of bribery.
To elaborate, the current system of manually monitoring traffic violations and collecting fines is not
only time-consuming but also prone to errors due to human involvement. Moreover, it is challenging to
track down and penalize offenders who try to evade fines or manipulate the system by bribing officials.
This, in turn, leads to an increase in the number of road accidents, which is a matter of grave concern.
Therefore, it is imperative to develop an automated system that can identify and penalize offenders
accurately and without delay.
The proposed system can incorporate the latest technology, such as artificial intelligence and machine
learning, to detect and capture traffic violations accurately and issue fines automatically. The system can
use cameras, sensors, and other devices to capture images and data of vehicles and drivers violating traffic
rules. The collected data can then be analyzed using AI algorithms to identify the type of violation and issue
an appropriate fine. By automating the process, the system can eliminate the possibility of human errors and
ensure transparency and accuracy in issuing fines.
III. LITERATURE SURVEY

In [1] Xuemei Xie research Says that The increasing illegal parking has become more and more serious.
Nowadays the methods of detecting illegally parked vehicles are based on background segmentation.
However, this method is weakly robust and sensitive to environment. Benefitting from deep learning, this
paper proposes a novel illegal vehicle parking detection system. Illegal vehicles captured by camera are
firstly located and classified by the famous Single Shot MultiBox Detector (SSD) algorithm. To improve the
performance, we propose to optimize SSD by adjusting the aspect ratio of default box to accommodate with
our dataset better.
In a research study conducted by Y Mohona Roopa [2], the primary goal was to fully automate the process of
detecting helmet usage and issuing fines. Previously, traffic surveillance was largely semi-automated or
completely reliant on human input. However, the new system being proposed aims to detect vehicles without
helmets using CCTV footage and extract the license plate. The number is then searched in a database to
obtain the registered phone number. The image of the footage, along with details of the fine and payment
options, is mailed and sent via message to the registered phone number. The information is also updated
online for future reference. This automated process allows for efficient and accurate enforcement of helmet
usage regulations and promotes compliance with traffic rules..
In [3] the proposed system is an android-based solution for automating the process of detecting and
enforcing traffic rule violations. It aims to address the challenges associated with manual traffic rule
enforcement by utilizing QR code technology and online payment methods.
In this system, a QR reader is used to scan the QR code attached to each vehicle. This code, which is
mandatory for all vehicles, is obtained when the vehicle is purchased or when the driver receives their
license from the Road Transport Office (RTO). When a violation is detected, the system extracts the QR
code from the footage and uses it to search for the registered phone number in a database. An image of
the footage, along with details of the violation and a payment request, is then sent to the registered phone
number via SMS or UPI app. The fine is automatically collected from the owner's bank account, and the
information is updated online for future reference.
In [4] The program for generating QR codes is implemented using the JavaScript language and run on a
node.js server. The generated QR code is displayed on a web browser for the user to view and scan. In
order to add an additional layer of security, the QR code is encrypted using Elliptical Curve Cryptography

(ECC).
To scan or read the encrypted QR code, the user needs to install a QR code scanner app on their
smartphone. There are many such apps available for free on app stores. Once the app is installed, the user
can start the application and use the smartphone camera to scan the QR code. The app will automatically
display the content of the QR code once it has been scanned.
In[5],First, a selection of QR code images were chosen and the mobile phone QR code reading software
was used to try and read the information contained within them. The statistical recognition rate of the QR
code images that could not be read by the software was then analyzed.
Next, the processed QR code images were read using the mobile phone QR code reading software, and the
statistical recognition rate was analyzed. The results showed that the QR code reading rate after processing
by the system algorithm in this article increased by 14%. This demonstrates the effectiveness of the
algorithm in improving the recognition rate of QR code images.
IV. PROPOSED METHODOLOGY
Fig1: flow diagram of proposed methodology

The proposed system aims to fully automate the process of detecting traffic rule violations and issuing
fines using CCTV cameras as shown in fig 1. When a violation is detected, the system extracts the vehicle's
license plate or QR code from the footage and searches for the associated phone number in a database. The
system then sends an image of the footage, along with details of the violation and a UPI payment request, to the
registeredphone number. The information is also updated online for future reference. The automated process
involves several stages, including the detection of the violation, identification of the vehicle, and issuance of the
fine..
A. Violation Recognition
In this module we are using Computer vision and Object detection technique to basically detect a vehicle
standing in non parking area, thereby capturing the QR code attached to the vehicle.Their will be a parking
violation.
B. Training over model

Training a model over Haar cascades involves using positive and negative samples of images to train a
classifier. Positive samples represent the object to be detected, while negative samples are images without the
object. These samples are used to train the classifier, which learns to distinguish between the object and non-
object images. The trained classifier can then be used to detect the object in new images or videos. Tuning the
parameters and the quality of the samples can improve the performance of the classifier.
C. QR Code Generation
In this module, we have used python’s qrcode module to generate the Qr Code and in this module encrypted data
is embedded inside the Qr code. We have used make_image module method of Qr code object to generate the
Qr code image.

D. QR Code Encryption and Decrption.

In this module we have intially enrypted the user’s information using Fernet symmetric key which is an
extension of AES algorithm. From this Fernet Algorithm a secret key is generated which is used for decrypt
users enrypted data during user info extraction.
E. Users Number Extraction

In this Module first the Qrcode is scanned when there is traffic violation. From the Qr code we extract the
users data after running fernet deryption algorithm on it. And from that Users data is send to backend server
through an exposed api endpoint.
F. Fine Generation
Once the user data is sent over server through an exposed API endpoint, the api module will fetch the data
with help of routing system in ExpressJs. These fetched data will be pushed to the database (i.e. MongoDB)
and simultaneously a payment request link will be send automatically through Email and SMS with the help
of instance.paymentLink.create method in Razorpay API.
All the details of rule violators will be displayed in tabular format on the E-challan website with help of MUI
library and ReactJs. The system also keeps track of the payment link status and user data in the database,
ensuring efficient and accurate record keeping. Whenever someone from officer side access the E-challan
website, the payment link staus of each violator is checked with help of
instance.paymentLink.fetch(req) in Razorpay APi and data in database will get updated. The data of users
who have paid the fine will be stored in different database from the data of users with pending status.
Users who have violated the rule can pay the fine with any method i.e.Debit/Credit card, Netbanking, or even
UPI. They will also get a message containing payment link, rule violated, amount to be paid which will be set
by default, and place where they were caught violating the rule.
Fig2. Activity in fine generatoin
G. Feedback Mechanism
Once the user pay the fine with payment link, they will also get a payment confirmation email from the
organisation involved. These acknowledgment will be sent with the help of Nodemailer module in NodeJs.
As the payment link status of the rule violator get updated (i.e. staus: paid), Nodemailer will send an
confirmation email tothe violator as a feedback mechanism.
IV Result Analysis
Traffic amercement automation is a system that automates the process of issuing and managing traffic fines.
The system uses various technologies like cameras, sensors, and artificial intelligence to capture traffic
violations and issue fines to the violators automatically.
The result analysis of traffic amercement automation includes various aspects like efficiency, accuracy, cost-

effectiveness, and safety.

Traffic amercement automation can significantly increase the efficiency of the traffic enforcement process.
The automated system can detect and issue fines to a large number of violators in a short period. This reduces
the workload on the traffic police and increases the efficiency of the traffic enforcement process.The
automated system is highly accurate and can detect traffic violations with a high degree of accuracy. This
reduces the chances of false fines and ensures that only genuine violators are fined.The system can
significantly improve safety on the roads by enforcing traffic rules strictly. It can also reduce the chances of
bribery and corruption, ensuring a fair and transparent system
However, there are also some concerns regarding the implementation of traffic amercement automation. Some
people argue that the system can be prone to errors, and there may be privacy concerns regarding the use of
cameras and sensors. The system may also face opposition from those who are fined and may lead to public
resentment if not implemented properly
Fig: Website of E-challan
Fig: fine created
Fig: Pyament requests

V Conclusion
Automating the process of collecting traffic fines using violation detection and qr code encryption can
have several benefits. One of the main advantages is that it can make the process more efficient and
convenient for both the authorities and the individuals who have received the fines. With QR code
recognition, authorities can easily identify vehicles that have been involved in traffic violations and
automatically send a fine to the owner of the vehicle by extracting data from Encrypted QR code. This can
save time and effort for both parties, as the individual does not have to physically visit a payment center to
pay their fine.
Another benefit is that automating the fine collection process can help to reduce errors and fraud. By using
violation detection technology, it is easier to track and verify transactions, ensuring that the correct amount
is being paid by the right person. This can help to prevent individuals from trying to evade paying their
fines or from being wrongly charged.
Overall, automating the process of collecting traffic fines for traffic violation and QR code technology can
improve the efficiency and accuracy of the process, making it more convenient for both the authorities and
individuals. It can also help to reduce errors and fraud, ensuring that the correct amount is being paid by the
right person.
Referance
[1] A S Mohammed Shariff Vehicle number plate detection using python and open cv. In 2021 Internation
Conference on Advance computing and Inovativetechnology I Engineering.
[2] Y Mohona Roopa & Sri Harishini Popuri Convulation Neural Network-based automatic extraction and
fine generation. In 2020 2nd International Conference on Inventive Research in Computing Application.
[3] S. Hariswetha, S. Indira A.& S. Latha., QR based Automatic Penalty Charging for Violation of Traffic
Rules and spark streaming. In 2020 International Journalof Engineering Research & Technology (IJERT)
[4] Vinit Kumar and Manish Kumar Dubey.Encrypted QR code generator. In 2019 International general of
recent scientific reaserch.
[5] Lina Hua, JIanxing zhu and Pradeep Kumar Singh. Qr image code recognition system based on AI
algorithm.In 2021 Journal of Intelligent System

Design and Traffic Ammercement

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Design and Traffic Ammercement

Uploaded by

Copyright:

Available Formats

VISVESVARAYA TECHNOLOGICAL UNIVERSITY

Chirag Suthar 1DT19CS033

Under the Guidance of:

Department of Computer Science and Engineering

Mrs. Arpitha Vasudev Dr. Kavitha C Dr. M. Ravishankar

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING

We, Chirag Suthar [1DT19CS033], Chirantan Banerjee [1DT19CS034], Gaurav Mourya

Sl. No. Name USN Signature with date

Chapter No. Page

3 Problem statement & proposed System 15

4 High Level Design 18

5 Low level Design 23

12 Copy of the Published paper Project Phase 1

13 Copy of the Published paper Project Phase 2

Figure No. FIGURE CAPTION Page No.

1.1 Computer Vision

 It will strictly upholds the traffic rules and discipline.

Dept. Of CSE, DSATM 2022-23 Page 8

1.3 Problem statement

Dept. Of CSE, DSATM 2022-23 Page 9

Dept. Of CSE, DSATM 2022-23 Page 10

2. Paper Title – VEHICLE NUMBER PLATE DETECTION USING PYTHON AND

Dept. Of CSE, DSATM 2022-23 Page 11

3. Paper Title: ENCRYPTED QR CODE GENERATOR

Dept. Of CSE, DSATM 2022-23 Page 12

4. Paper Title: Convolutional Neural Network-based Automatic Extraction and Fine

Dept. Of CSE, DSATM 2022-23 Page 13

Research in traffic management automation is crucial to advancing the development and

Dept. Of CSE, DSATM 2022-23 Page 14

PROBLEM STATEMENT AND PROPOSED SYSTEM

3.1 Problem Statement

3.2 Proposed System

Dept. Of CSE, DSATM 2022-23 Page 15

3.3 Hardware Requirements

3.4 Software Requirements

Dept. Of CSE, DSATM 2022-23 Page 16

Dept. Of CSE, DSATM 2022-23 Page 17

4.1 System Architecture

Dept. Of CSE, DSATM 2022-23 Page 18

4.2 System Design

Figure 4.1: Level 0 Data Flow Diagram

Dept. Of CSE, DSATM 2022-23 Page 19

Figure 4.2: Level 1 Data Flow Diagram

4.3 Modules identified.

This project is mainly divided into 4 modules :

Dept. Of CSE, DSATM 2022-23 Page 20

QR code generation is the process of creating QR codes, which are two-dimensional

3. Users Information Extraction:

Fernet encryption is a symmetric encryption algorithm that generates a secure

The purpose of a feedback mechanism is to help organizations or individuals identify

Dept. Of CSE, DSATM 2022-23 Page 21

Dept. Of CSE, DSATM 2022-23 Page 22

LOW LEVEL DESIGN

5.1 Algorithms Used

5.1.1 Haar Cascade

Dept. Of CSE, DSATM 2022-23 Page 23

5.1.2 Fernet Algorithm :

Dept. Of CSE, DSATM 2022-23 Page 24

Figure 5.1: Methodology flow diagram

Dept. Of CSE, DSATM 2022-23 Page 25

6.2 Module1. Qr Code Generation and Encryption

Dept. Of CSE, DSATM 2022-23 Page 26

6.4 Module 3. Fine Generation