Volume 8, Issue 6, June 2023 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165

Traffic Rules Violation Detection System

Hirnaik Ashlesha Padmakar, Malvanakar Gururaj Pradeep, Rakshe Omkar Ravindra, Ahire Hrishikesh Sachin,
Prof. Vidya Gogate, Prof. Dr. Sarika Bukkawar
Department of Electronics Engineering
Shahand Anchor Kutchhi Engineering College
Mumbai: 400088, India

Abstract:- This paper represents design and Traffic rules violation detection systems is promising
implementation of Traffic Rules Violation Detection system, as new advancements in technology and artificial
System. The traffic rules violation detection system is an intelligence continue to be developed.
advanced technology that can identify and analyze traffic
violations using computer vision techniques. The system II. LITERATURE SURVEY
uses cameras installed at traffic junctions to capture
images and videos of vehicles violating traffic rules. 1. "A Survey of Intelligent Traffic Systems" by S.
These images and videos are then processed using Pandey and S. K. Singh (2021)provides a comprehensive
artificial intelligence algorithms to detect and classify the survey of intelligent traffic systems, including traffic
type of violation, such as jumping a red light, over- violation detection systems. The authors discuss the
speeding, or lane violation. The system can also generate various types of traffic violations, the challenges in
automated alerts and notifications to law enforcement detecting and classifying them, and the techniques used for
agencies to take action against the violators. The system object detection and classification. They also highlight the
helps in reducing the number of accidents and improving limitations of current systems and suggest areas for further
overall road safety by enforcing traffic rules and research, such as the use of multimodal data and the
regulations. integration of real-time data sources.[7]"An Overview of
Traffic Violation Detection and Monitoring Systems" by S.
Keywords:- Python,OpenCV, Image Processing. Ghose, S. Sarkar, and S. Das (2021) provides an overview
of traffic violation detection and monitoring systems and
I. INTRODUCTION discusses the various techniques used for object detection
and classification. The authors also discuss the challenges
The traffic rules violation detection system is an in implementing such systems, such as the need for
advanced technology that has been developed to improve accurate data and the difficulty in detecting violations in
road safety by enforcing traffic rules and regulations. The complex traffic situations. They suggest the need for
system uses computer vision techniques to capture images continued research in this field to improve the accuracy and
and videos of vehicles violating traffic rules, such as over- efficiency of traffic violation detection systems. [8] "Real-
speeding, jumping red lights, or lane violations. The system Time Traffic Rule Violation Detection and Alert System
is equipped with artificial intelligence algorithms that can using Deep Learning" by R. Gupta and P. Anand
detect and classify the type of violation and generate (2020)proposes a real-time traffic rule violation detection
automated alerts and notifications to law enforcement and alert system using deep learning techniques. The
agencies. authors use a YOLOv3 algorithm for object detection and
classification and an LSTM network for real-time alert
The need for such a system arises from the increasing generation. The paper also discusses the experimental
number of accidents and fatalities caused by reckless driving results and highlights the system's accuracy and
and violation of traffic rules. According to the World Health efficiency.[9] "Traffic Rule Violation Detection System
Organization, road accidents are the eighth leading cause of using Deep Learning Techniques" by R. Jain and V. Jain
death globally, with approximately 1.35 million people (2019)proposes a traffic rule violation detection system
losing their lives every year. In addition to the human cost, based on deep learning techniques. The authors use a
road accidents also result in significant economic losses due YOLOv3 algorithm for object detection and classification
to damage to vehicles and infrastructure. and an SVM classifier for traffic rule violation detection.
To address these issues, governments and law The paper also discusses the experimental results and
enforcement agencies have been implementing various highlights the system's accuracy and efficiency. [10]
measures, including stricter traffic laws and increased Overall, these papers provide a detailed overview of
enforcement. The traffic rules violation detection system is a traffic violation detection systems, highlighting the
technological solution that can complement these efforts by challenges and limitations of current systems and
providing an automated and efficient method of detecting suggesting areas for further research. They also propose
and reporting traffic violations. novel techniques for object detection and classification and
demonstrate the effectiveness of deep learning techniques
in improving the accuracy and efficiency of traffic
violation detection systems.

IJISRT23JUN764 www.ijisrt.com 1582

Volume 8, Issue 6, June 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
III. SYSTEM MODEL

A. Activity Diagram
The Activity diagram is another important behavioral diagram in UML diagram to describe dynamic aspects of the system.
They describe how the activities are coordinated to provide a service which can be at different levels of abstraction.

Fig. 1: Activity Diagram of the system

B. Flowchart Of System:

Fig. 2: Flowchart Of System

IJISRT23JUN764 www.ijisrt.com 1583

Volume 8, Issue 6, June 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
C. FUNCTIONAL MODELLING:
With the aid of DFDs (Data Flow Diagram), we will define the function of the internal processes of the system. This
perspective concentrates on describing the dynamic process of our deepfake detection system.

 LEVEL 0:

Fig. 3: Level 0 Data Flow Diagram

The “Context level” DFD is the basic overview of the whole system. The user uploads the video, based on which, our system
accordingly processes and gives the output to the user. If found to be deepfake, the video is saved to our system.

 LEVEL 1:

Fig. 4: Level 1 Data Flow Diagram

The Context diagram is decomposed into multiple processes. The main functions of the system are highlighted, where the
high-level processes are broken down into subprocesses.

 LEVEL 2:

Fig. 5: Level 2 Data Flow Diagram

The level 2 DFD offers a more detailed look at the each video is checked frame by frame. If found to be
processes that make up an information system. The diagram genuine, the results are displayed to the user. If found fake,
defines all the processes and subprocesses of the system the video is added to the central database with a “Deep
through which the user gets the required output. Fake” watermark which is then shared to the user to
download or share it to others.
In short, the user requests the validation of an
image/video, which is then processed by the system where

IJISRT23JUN764 www.ijisrt.com 1584

Volume 8, Issue 6, June 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
D. TIMELINE CHART:

(a)

(b)
Fig. 6: Timeline chart of our Project

E. ARCHITECTURAL DESIGN

Fig. 7: Architectural Design

With the help of the user interface of our application, leave their feedback.
users can upload their videos to the system which will then
classify it as genuine or Deepfake.Along with that the The classification system will also be available
system will flag the frames that have been altered and label through the system’s API enabling offsite users to
the video as fake on the whole. implement this tool in their application/system the way
they want, for which the result will be sent back to the
If given the proper permissions to upload the video to system.
the central database, other users can also download the same
video. For our deep learning model, the system takes the
uploaded input from the user which is then passed on to our
Hence, all flagged fake videos are available for public trained model. The results are again classified into real and
access, where visitors can also view recent or fake videos, which are then passed on to the user.
frequent/popular classified Deep Fakes on the website and

IJISRT23JUN764 www.ijisrt.com 1585

Volume 8, Issue 6, June 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
The objective in flagging and marking the video,
making it available for download is an effort to spread the
classified video among networks and popularize it just the
way Deep Fakes are spread and popularized so there is a
crackdown on it’s misuse and malpractice.

IV. IMPLEMENTATIONS

A. Algorithms with implementations:

 Image Preprocessing: The first step is to convert the
colored image into grayscale. This is done to reduce the
complexity of the image and make it easier to detect edges
using the Canny edge detection algorithm. The grayscale Fig. 10: Rule Violation Dection
image is then filtered using a Gaussian filter to reduce
noise. V. FUTURE SCOPE
 Edge Detection: The Canny edge detection algorithm is
then used to identify the edges of the image. The Canny The future scope of traffic rules violation detection
algorithm works by identifying the areas of the image systems is promising, as new advancements in technology
where there is a sharp change in brightness, indicating an and artificial intelligence continue to be developed. Some
edge. This algorithm produces a binary image where the potential areas of growth and improvement include:
edges are represented as white pixels and the non-edges as  Integration with Autonomous Vehicles: As
black pixels. autonomous vehicles become more common on roads,
 Line Detection: Once the edges have been detected, the traffic rules violation detection systems could be
Hough transform is used to identify the lines in the image. integrated with these vehicles to ensure that they adhere
The Hough transform works by converting the binary to traffic rules and regulations.
image into a parameter space where each point represents  Real-time Monitoring: The development of faster and
a line in the original image. The lines are identified by more efficient data processing techniques could enable
finding the peaks in the parameter space. real-time monitoring of traffic violations, allowing for
 Rule Violation Detection: Once the lines have been quicker enforcement and response times.
detected, they can be analyzed to detect traffic rule  Improved Accuracy: The use of more advanced
violations. For example, if the image is of a road and the algorithms and machine learning models could lead to
lines represent the lane markings, the distance between the improved accuracy in detecting and classifying different
lines can be analyzed to detect if a vehicle is driving types of traffic violations.
outside its lane.  Enhanced Data Analytics: The data collected by traffic
rules violation detection systems can be used to analyze
traffic patterns and identify areas of high traffic violation
rates. This information can then be used to develop more
effective traffic management strategies.
 Expansion to New Regions: Traffic rules violation
detection systems can be implemented in new regions to
improve road safety and reduce the number of accidents
caused by reckless driving.
 Integration with Smart City Initiatives: The
integration of traffic rules violation detection systems
with other smart city initiatives, such as smart traffic
management systems and public transportation systems,
can lead to a more efficient and sustainable
Fig. 8: Edge Detection transportation system.

Overall, the future scope of traffic rules violation

detection systems is promising, and continued
advancements in technology and artificial intelligence are
likely to lead to even more effective and efficient systems
in the years to come. The ESP32-CAM is a cost-effective
development board based on the ESP32 microcontroller,
featuring an onboard camera and compact size. Designed
for IoT applications, prototype constructions, and DIY
projects, this board integrates Wi-Fi, traditional Bluetooth,
and low-power BLE, powered by two high-performance
32-bit LX6 CPUs.
Fig. 9: Line Detection

IJISRT23JUN764 www.ijisrt.com 1586

Volume 8, Issue 6, June 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
VI. RESULTS AND ANALYSIS Applications, 79(17), 12073-12094.
[4.] Nguyen, T. T., Nguyen, H. T., Nguyen, T. T. T., &
Firstly, for beginning the project, the user requires to Vu, Q. D. (2021). A real-time traffic rule violation
open a video using “Open” option that is visible under file. detection and recognition system based on deep
The user can open any footage from the system storage. learning. IEEE Access, 9, 50153-50163.
After opening it, the system will load a preview. It contains [5.] Khursheed, O., Saleem, A., & Qureshi, M. A.
one of the frames of the input video. It is used to recognize (2020). Smart traffic rule violation detection and
the roads and select region of interest. The selection by the notification system using deep learning.
user will act as a traffic line. To draw this line, we need to International Journal of Advanced Computer
click on “Region of interest” item from the Analyze option. Science and Applications, 11(7), 204-210.
After clicking on it, the user needs to pick two points to plot [6.] Mehmood, A., Javed, A. R., Baig, A. R., & Akram,
a line that indicates the traffic signal. Selection of the traffic A. (2021). Traffic Rule Violation Detection System
line will automate the process of violation detection. Now Using Deep Learning: A Review. In Advances in
the system identifies the vehicles and then it is ready to Computer Science for Engineering and Education II
detect the violations. After that, GUI gives output frame by (pp. 1-12). Springer, Cham.
frame. The system continues to display the result until it [7.] Khanna, R., Agrawal, S., & Sharma, S. (2018).
reaches the final frame of the video footage. In the Automatic Traffic Rule Violation Detection and
background, an output file will be generated. This file can be Reporting System. International Journal of
found in output folder of Resources. This process of Computer Sciences and Engineering, 6(5), 87-90.
violation detection will be terminated. After applying the [8.] "A Survey of Intelligent Traffic Systems" by S.
process to all the frames of the input video footage, the user Pandey and S. K. Singh (2021)
can give another video as input.If the work is completed the [9.] "An Overview of Traffic Violation Detection and
user can quit using Exit item available in the File option. Monitoring Systems" by S. Ghose, S. Sarkar, and S.
Das (2021)
VII. CONCLUSION [10.] "Real-Time Traffic Rule Violation Detection and
A traffic rules violation detection system is an Alert System using Deep Learning" by R. Gupta and
advanced technology that has the potential to significantly P. Anand (2020)
improve road safety and traffic management. By using a [11.] "Traffic Rule Violation Detection System using
combination of cameras, computer vision algorithms, and Deep Learning Techniques" by R. Jain and V. Jain
machine learning models, the system can detect and classify (2019)
various types of traffic violations, such as running red lights,
speeding, and illegal parking. The system can also generate
automated alerts and notifications for law enforcement
agencies, allowing for quicker enforcement and response
times.

Implementing a traffic rules violation detection system

can lead to reduced accidents, improved compliance with
traffic rules and regulations, increased efficiency, better data
collection, and improved public safety. The future scope of
such systems is also promising, with potential integration
with autonomous vehicles, real-time monitoring, and
enhanced accuracy in detecting and classifying traffic
violations.

Overall, a traffic rules violation detection system is a

valuable tool for improving road safety and traffic
management, and its continued development and
implementation can lead to significant positive impacts on
communities and society as a whole.

REFERENCES

[1.] Here are some references related to traffic rules

violation detection systems:
[2.] Liu, C., Xu, M., Wang, W., & Zhou, X. (2020). A
Review of Intelligent Transportation System Based
on Traffic Violation Detection. Journal of Advanced
Transportation, 2020, 1-16.
[3.] Singh, K. K., & Kumar, V. (2020). A novel intelligent
traffic rule violation detection and management
system using deep learning. Multimedia Tools and

IJISRT23JUN764 www.ijisrt.com 1587