Table of Contents

Abstract .......................................................................................................................................... …
Chapter 1: Introduction…….…………………………..…………………………………….……..
1.1 Background……………..……………………………………………………………….………
1.2 Statement of the Problem……………………………………………………………….……….
1.3 Research Aims and Objectives……….………………………………………………….……...
1.3.1 Research Aims………………………………………………………………………...
1.3.2 Research Objectives…………………………………………………………………..
1.4 Why it is Well-Motivated………………………...…………….……………………………….
1.5 Significance of the Study…………………………………………………………….………….
1.6 Scope and Limitations…………………………………………………………………………..
1.6.1 Scope………………………………………………………………………………….
1.6.2 Limitations…………………………………………………………………………….
1.7 Thesis Structure……..…………………………………………………………………………..
ABSTRACT
The Gambia has had traffic congestion and road safety issues for years. The first problem is
inadequate roads and poor traffic control. Road accidents are the order of the day. Traffic
congestion is one of the crucial complications in urban cities in The Gambia. This needs to be
addressed to improve traffic flow, control, and operation. The ever-increasing nature makes it
imperative to know the road traffic density in real time for better signal control and effective
traffic management. There can be different causes of congestion in traffic like insufficient
capacity, unrestrained demand, large Red-Light delays, etc. This project studies the reports of the
use of smart technologies and Artificial Intelligence (AI) to improve road safety while ushering
in a new era of traffic management.
While insufficient capacity and unrestrained demand are somewhere interrelated, the delay of
respective light is hard coded and not dependent on traffic. Therefore, the need for simulating
and optimizing traffic control to better accommodate this increasing demand arises. In recent
years, image processing and surveillance systems have been widely used in traffic
management for traveler information, ramp metering, and updates in real-time. The traffic
density estimation can also be achieved using Image Processing. This project presents the
method to use live image feed from the cameras at traffic junctions for real-time traffic density
calculation using image processing. This system employs YOLOv5 and AlexNet V3
Convolutional neural network pre-trained model to accurately detect the number of vehicles
present on the road and average Vehicle Area and identify emergency vehicles in real-time. It
also focuses on the algorithm for switching the traffic lights according to vehicle density on the
road, thereby aiming at reducing the traffic congestion on roads which will help lower the
number of accidents. In turn, it will provide safe transit to people and reduce fuel
consumption and waiting time. It will also provide significant data which will help in future
road planning and analysis. In further stages, multiple traffic lights can be synchronized with
each other with the aim of even less traffic congestion and free flow of traffic. The vehicles
are detected by the system through images instead of using electronic sensors embedded in the
pavement. A camera will be placed alongside the traffic light to capture image sequences. Image
processing is a better technique to control the state change of the traffic light. It shows that it
can decrease traffic congestion and avoid the time being wasted by a green light on an empty
road. It is also more reliable in estimating vehicle presence because it uses actual traffic images.
It visualizes the practicality, so it functions much better than those systems that rely on the
detection of the vehicles’ metal content. The system is essential for efficient traffic management,
and this system uses complex algorithms to reliably detect and monitor several lanes, including
emergency lanes. It also uses pedestrian detection techniques to make pedestrian crossings safer
and more efficient. To complete this project, various technological tools were used such as
Python, OpenCV libraries, and Anaconda IDE.
Keywords: Road safety, traffic congestion, smart technologies, traffic lights, traffic management,
image processing, YOLOv5, Alexnet V3 Convolutional Neural Network.

2
1. Introduction

Urban transport problems are a worldwide problem. The Gambia, like many other developing
countries, faces numerous challenges in traffic management and road safety. These challenges
include inadequate infrastructure, lack of traffic monitoring systems, and limited resources for
effective traffic management. The country has poor roads. Due to this, traffic accidents and
congestion are very common. Urban communities have a notable issue of fixed time cycles for
traffic signals at crossing points. Small activities like stopping a vehicle at an intersection or a
vehicle breaking the traffic signal cause a chain reaction which at last prompts colossal car
influxes. This is backed by a study done by Dr. Gabor Orosz of the University of Exeter [1].

Effective traffic management is important to minimize congestion, protect individuals on the

road, and keep the road safe. Traditional traffic management systems often struggle to cope with
increasing congestion, leading to longer commute times, heightened fuel consumption, and
elevated pollution levels. Moreover, the limitations of manual monitoring and human intervention
have underscored the need for innovative approaches to address these pressing issues.
The framework that is currently in application uses a predetermined time duration for the traffic
signal and any changes require human intervention and manual traffic control by the policemen.
This system is highly inefficient as the person violating the traffic signal needs to be caught by
the traffic police physically which leads to further jamming of lanes [2]. Let us look at how use of
Information and data available to us can be used to create some smart services, which in a true
sense provide us with a better living. A better approach will be to decide the signal time of the
number of cars crossing the signal. If there are a greater number of cars on a particular lane, the
signal time will be increased to clear congestion.
This paper aims to do a comprehensive study that explores the integration of cutting-edge smart
technologies with data-driven insights to revolutionize traffic management and improve road
safety. We have achieved this with the help of image processing that can be obtained from
surveillance cameras and eventually to deploy a feedback mechanism in the working of the
traffic lights where the density of the traffic would also be factored in the decision-making
process. The document discusses the development of a smart traffic management system that
utilizes YOLOv5 (You Only Look Once) and AlexNetV3 Convolutional Neural Network
Architecture to optimize traffic flow, improve emergency response times, and address the
challenges of existing traffic control systems.

1.1 Background of Road Safety and Traffic Management System in The Gambia
According to statistics, the Gambia globally ranked among the 4th countries with the highest
number of road accidents. Based on the statistics obtained from the police from the year 2011 to
2021, about 121 people die annually as a result of road traffic accidents in the Gambia. The

3
situation seems to be worsening looking at the statistics for 2021, when a total of 1218 road
crashes occurred, with 155 people deaths and 440 people sustaining serious injuries.
It is estimated that each year, road traffic deaths and injuries cost the Gambia over two billion
dalasi through treatment costs, loss of productivity, pain, and ongoing suffering. This was
revealed by Ebrima Colley, Deputy Permanent Secretary-Technical at the Ministry of Transport,
Works and Infrastructure, during a day-long sensitization workshop on road safety for National
Assembly Members held on Tuesday, 27th December 2022 at the Senegambia Beach Hotel [3].
The Gambia's road network is relatively underdeveloped, with many roads in poor condition and
inadequate traffic management systems. Additionally, the country has a high number of second-
hand vehicles, which may be poorly maintained and contribute to the high rate of accidents [4].
Road traffic accidents occurred as a result of faulty driving during the period under review. This
suggests that most drivers are either careless or unqualified. This could be affiliated with the
inappropriate issuance of driving licenses. To reduce road traffic accidents, there is a need to
establish well-equipped driving schools.
The Gambia, characterized by its vibrant communities and diverse road networks, faces
multifaceted challenges in the realm of road safety and traffic management. Rapid urbanization,
coupled with the complexities of a growing population, demands innovative strategies that extend
beyond traditional approaches. The pressing need for a comprehensive and adaptive system to
address the unique nuances of Gambian roads propels this research into the forefront of AI-driven
interventions.
The challenges posed by urban traffic congestion and the need for efficient traffic control and
operation. Traffic congestion is a common issue in urban areas, leading to delays, increased travel
times, economic costs, and environmental impacts. To tackle these challenges, various
approaches and technologies have been developed to optimize traffic flow and improve
emergency response times.

1.2 Statement of the Problem

While several countries in the region have adopted policies to encourage walking and cycling,
there is concern that if these are not accompanied by additional safety measures – notably speed
reduction, they will not achieve the desired effects, and may indeed make roads more dangerous
for vulnerable road users [6]. The Gambia, like many developing nations, grapples with
significant challenges in the realms of traffic management and road safety. Rapid urbanization,
coupled with a burgeoning population and diverse road networks, has intensified the complexities
of ensuring efficient traffic flow and minimizing road accidents. Traditional approaches to traffic
management often fall short of addressing the dynamic and multifaceted nature of these
challenges.

4
Current road safety measures face limitations in timely accident detection, adaptive traffic signal
control, and optimized emergency response, contributing to persistent hazards on Gambian roads.
The absence of a comprehensive and technologically advanced system leaves the road
infrastructure susceptible to congestion, accidents, and delays in emergency services. This critical
gap in road safety measures necessitates a paradigm shift, calling for innovative solutions that can
adapt to the unique dynamics of Gambian roads.
Moreover, the lack of localized and context-aware solutions further exacerbates the road safety
dilemma. Generic traffic management systems, bereft of an understanding of local nuances, fail
to provide tailored interventions that resonate with the specific challenges and needs of The
Gambia.
In light of these pressing concerns, this project identifies a critical need for a cutting-edge
approach to traffic management and road safety enhancement, specifically through the
incorporation of Artificial Intelligence (AI) solutions. The complexities of Gambian roads
demand a sophisticated and adaptive system that harnesses the power of AI to predict, prevent,
and respond to traffic-related challenges in real-time.
The overarching problem, therefore, lies in the absence of a comprehensive, context-aware, and
technologically advanced approach to traffic management and road safety in The Gambia. This
project aims to fill this void by investigating and implementing AI solutions that not only align
with the unique characteristics of Gambian roads but also set a precedent for the integration of
technology to enhance road safety in developing nations.

1.3 Research Aim and Objectives

1.3.1. Research Aims
1. To Develop AI-Driven Traffic Management Solutions:

The paper’s objective is to develop advanced traffic management solutions tailored to The
Gambia's unique traffic dynamics. This will involve exploring and creating artificial intelligence
algorithms and machine learning techniques, specifically focusing on the use of Convolutional
Neural Networks (CNN) such as YOLOv5. The ultimate aim is to enhance real-time traffic flow
and alleviate congestion by leveraging cutting-edge image processing techniques. The study aims
to contribute to the existing body of knowledge in traffic management and the application of
artificial intelligence in addressing traffic challenges, with a specific focus on the context of The
Gambia.

2. To Enhance Road Safety through AI Applications:

5
The paper will focus on the implementation of cutting-edge computer vision technologies to
detect accidents in real-time, with the aim of enhancing road safety and reducing the incidence of
accidents on Gambian roads. An important aspect of this study will be the exploration of the
application of pre-trained Convolutional Neural Network models to improve the precision and
accuracy of vehicle detection. By leveraging these advanced technological tools, it is anticipated
that the paper will contribute significantly to the development of effective traffic management
solutions tailored to the unique traffic dynamics of The Gambia. This approach aligns with the
broader objective of leveraging artificial intelligence and machine learning techniques to address
traffic challenges. Moreover, the study will involve comprehensive data collection and analysis
of traffic patterns, road infrastructure, and driver behavior to provide valuable insights for
enhancing traffic control and management in The Gambia.

3. To Optimize Emergency Response Times:

The utilization of YOLOv5-based predictive models can improve emergency response routes in
real-time traffic conditions, leading to swift and effective services for accidents and emergencies.
This will have a positive impact on road safety and accident rates. Through precise monitoring of
traffic flow, alleviation of congestion, and the application of advanced computer vision
technologies alongside intelligent traffic control measures, it is possible to establish a safer
transportation system that benefits both residents and visitors. Implementation of these strategies
not only enhances road safety but also contributes to the overall development of transportation
infrastructure in The Gambia while improving the quality of life for its citizens. The research
seeks to enhance tailored traffic control solutions suited to The Gambia's unique patterns by
analyzing traffic flows, utilizing artificial intelligence algorithms, and developing customized
approaches aimed at accurately monitoring traffic movement, decreasing congestion, as well as
augmenting road safety through sophisticated computer vision technologies coupled with
intelligent traffic control measures - thus creating a safer and more efficient transportation system
in The Gambia.

4. Real-Time Traffic Monitoring and Analysis

The paper seeks to examine traffic flow, road infrastructure, and driver conduct in The Gambia.
By employing AI algorithms such as YOLOv5 by leveraging its speed and AlexNet V3’s
accuracy, the goal is to create a thorough system for monitoring and analyzing real-time traffic.
This will result in a prompt understanding of vehicle movements on roads and ultimately improve
traffic regulation and governance in The Gambia. Incorporating predictive models based on
YOLOv5 will optimize emergency response routes under live traffic conditions, ensuring rapid
assistance for accidents and emergencies - thereby contributing to improved road safety.
Furthermore, intelligent traffic management systems utilizing computer vision technologies will
enable pre-emptive identification and mitigation of potential risks on the roads. Ultimately, this
research aims to transform transportation in The Gambia by harnessing advanced technologies
for a safer, more efficient, and sustainable transportation system that benefits all stakeholders.

6
 5. Evaluate and Adapt AI Solutions to Local Contexts:

Assessing the effectiveness and versatility of AI solutions like AlexNet V3, CNN, and YOLOv5
in The Gambia requires an examination of local traffic patterns, existing infrastructure, and
emergency service capabilities. This evaluation will uncover how these AI algorithms can be
utilized to analyze and tackle traffic issues in The Gambia, ultimately enhancing road safety,
traffic flow, and emergency response. Integration of these AI solutions into The Gambia's
transportation system is anticipated to significantly diminish road accidents and fatalities.
Additionally, AI-powered traffic management systems can streamline traffic flow, reduce
congestion, and improve overall transportation efficiency in The Gambia while also delivering
economic benefits such as cost savings through decreased delays and enhanced fuel efficiency.
This paper seeks to drive the digital transformation of transportation in The Gambia by equipping
the country with advanced AI technologies that foster a safer, more efficient transport system.

6. Performance Evaluation and Optimization:

By thoroughly evaluating the combined YOLOv5 and AlexNet V3 systems, we can refine their
parameters and algorithms to ensure effectiveness, precision, and expandability in the vehicle
detection solution. This is essential for enabling real-time traffic monitoring and management in
The Gambia, ultimately leading to enhanced road safety, reduced congestion, and improved
transportation efficiency. The adoption of advanced AI technologies like YOLOv5 and AlexNet
V3 holds the potential to modernize The Gambia's transportation system by improving road
safety, traffic flow, and emergency response capabilities for all stakeholders.

7. Adaptive Traffic Management Solutions:

Implementing intelligent traffic management solutions in The Gambia involves utilizing insights
driven by AI to devise strategies for improving traffic flow, reducing congestion, and enhancing
overall transportation efficiency. This effort holds the potential to significantly boost the
effectiveness of the transport system in The Gambia by incorporating advanced AI technologies
into traffic management. The country can anticipate a decline in road accidents and casualties, as
well as economic benefits from cost savings due to reduced delays and improved fuel efficiency.
Implementing adaptive traffic management solutions based on AI insights offers prospects for
spearheading the digital transformation of transportation in The Gambia by establishing a safer,
more efficient, and sustainable transportation network that serves all stakeholders' interests.
Overall, leveraging AI technologies and adaptive traffic management solutions has promising
potential to modernize The Gambia's transportation system through improved road safety and
optimized transport efficiency amid today's rapidly evolving environment. However, it is
important to note that the success of implementing AI-driven traffic management solutions in The
Gambia will depend on factors such as investment in infrastructure, regulatory framework, and
public awareness and acceptance of these technologies.
7
 8. Integration of Neural Network Models for Synergistic Impact:
Investigate the synergistic impact of integrating and exploring neural network models to elevate
the effectiveness of vehicle detection in diverse and challenging traffic scenarios.
PARAPHRASED

By incorporating and examining neural network models, the research intends to evaluate the
potential enhancements in vehicle detection within varied and complex traffic conditions by
utilizing AI technologies such as YOLOv5 and AlexNet V3. The objective is to optimize the
parameters and algorithms of these combined systems to guarantee their accuracy, efficiency, and
flexibility for immediate traffic supervision and control in The Gambia. Ultimately, the study
aims to advance digital innovation in transportation by establishing safer, more streamlined, and
environmentally conscious practices through advanced AI-powered solutions for traffic
management.

1.3.2 Research Objectives

1. Integration of AlexNet V3 for Feature Enhancement

Incorporating the AlexNet V3 model into the vehicle detection system represents a crucial
advancement in feature extraction, classification accuracy improvement, and enhancement of
intricate vehicle pattern identification within diverse and challenging traffic scenarios. By
integrating AlexNet V3, the research seeks to leverage its deep learning capabilities to improve
recognition and understanding of complex vehicle features, resulting in more precise and efficient
detection. This integration is expected to enhance the overall effectiveness and reliability of the
vehicle detection system, ultimately contributing to modernizing and increasing transportation
efficiency in The Gambia. Additionally, implementing AlexNet V3 will fortify the adaptability
and scalability of the vehicle detection solution, setting a foundation for real-time monitoring and
traffic flow management.

2. Integrate YOLOv5 for Real-Time Accident Detection:

Integrate YOLOv5, a state-of-the-art object detection model, for real-time accident detection,
utilizing CNN and image processing to analyze live camera feeds and promptly identify and
report accidents or potential hazards.

3. Create Predictive Models for Emergency Response Optimization:

8
Develop and implement YOLOv5-based predictive models to optimize emergency response
routes based on real-time traffic conditions, aiming to minimize response times and enhance the
effectiveness of emergency services.
4. Evaluate CNN and YOLOv5 Performance in Gambian Context:

By assessing the application of CNN and YOLOv5-based AI technologies in the context of traffic
and infrastructure management in The Gambia, valuable insights can be obtained regarding their
effectiveness, efficiency, and suitability for enhancing traffic and infrastructure control. The
findings from this assessment could guide future decision-making and resource allocation in The
Gambia, potentially leading to the formulation of more efficient strategies for managing traffic
and infrastructure. Implementing CNN and YOLOv5-based AI solutions within the framework of
traffic and infrastructure in The Gambia offers an opportunity to gain valuable insights into how
these AI solutions can enhance the management of traffic flow as well as infrastructural systems.

5. Creation of a Dynamic Traffic Monitoring System:

We plan to create a sophisticated system that constantly monitors and assesses the movements of
vehicles in real-time, providing a thorough examination of traffic trends. Through the application
of artificial intelligence algorithms and machine learning methods, the system will detect areas
with heavy traffic, forecast traffic patterns, and recommend the most efficient routes for drivers.

1.4 Why it is well-motivated

The project is well-motivated due to its innovative technological approach, the specificity of
addressing local challenges, the potential for immediate impact, and the combination of advanced
AI models with practical applications. It aligns with the broader goals of improving road safety
and traffic management while incorporating state-of-the-art technologies tailored to the unique
context of The Gambia.
1. Uniqueness of Technological Integration

The integration of cutting-edge technologies, such as Convolutional Neural Networks (CNN) and
YOLOv5, into the context of traffic management and road safety in The Gambia, is a novel,
groundbreaking, unparalleled, and unique approach. This research is motivated by the potential
transformative impact of these technologies on addressing specific challenges encountered by
Gambian roads.

2. Addressing Specific Local Challenges

9
The focus on The Gambia ensures that the research is highly context-specific, considering the
unique dynamics, infrastructure, and traffic patterns of the country. By tailoring AI solutions to
the local context, the research addresses challenges that are directly relevant to The Gambia,
contributing to more effective and targeted interventions.

3. Potential for Real-Time Impact

The emphasis on real-time applications, such as real-time traffic flow optimization, accident
detection, and emergency response optimization, underscores the potential for immediate impact.
This research is motivated by the prospect of implementing solutions that can be deployed
rapidly to enhance road safety and traffic management in the short term.
4. Advanced Object Detection with YOLOv5
The use of YOLOv5, a state-of-the-art object detection model, for real-time accident detection
and image analysis is a key motivator. YOLOv5's capabilities in efficient and accurate object
detection contribute to the research's potential for achieving high-performance results in critical
areas of road safety.
5. Smart City and Infrastructure Planning
Beyond immediate applications, the project holds significance in the broader context of smart
city initiatives. The data generated from YOLOv5 and AlexNet V3 can inform city planners and
authorities in making data-driven decisions for future infrastructure enhancements, traffic
planning, and urban development.

1.5 Significance of the Study

The significance of the study lies in its potential to bring about transformative changes in the
realm of traffic management and road safety in The Gambia. The outcomes of the research can
contribute not only to the improvement of transportation systems in The Gambia but also to the
broader field of AI applications in road safety globally. The key aspects that highlight the
significance of the study:
1. Improved Road Safety
Safety is a paramount concern in traffic management, and the project's significance lies in its
potential to enhance safety measures on roadways. The study aims to enhance road safety by
employing cutting-edge technologies such as Convolutional Neural Networks (CNN) and
YOLOv5 for real-time accident detection. This has the potential to significantly reduce the
occurrence of accidents and mitigate their severity, thereby safeguarding the lives of both drivers
and pedestrians.

10
2. Enhanced Traffic Management Efficiency:
By developing and implementing AI-driven traffic management solutions, including real-time
traffic flow optimization, the study seeks to improve the efficiency of traffic management in The
Gambia. This can result in reduced congestion, smoother traffic flow, and an overall
improvement in the transportation infrastructure.
3. Swift Emergency Response:
The integration of YOLOv5-based predictive models aims to optimize emergency response
routes based on real-time traffic conditions. This can lead to faster response times during
accidents or emergencies, potentially saving lives and minimizing the impact of critical
situations.
4. Tailored Solutions for Local Context:
The significance of the study lies in its focus on developing and adapting AI solutions to the
unique context of The Gambia. This ensures that the proposed interventions are specifically
tailored to address the challenges and characteristics of Gambian roads, making the research
directly relevant to the local population.
5. Adoption of State-of-the-Art Technologies:
The study's incorporation of advanced technologies such as YOLOv5 and CNN positions it at the
forefront of technological innovation in the field of transportation. The adoption of state-of-the-
art models for real-time applications showcases the potential for leveraging the latest
advancements in AI for practical and impactful solutions.
6. Scalability and Replicability:
The study's focus on creating a model for traffic management and road safety that integrates AI
technologies has broader implications. If successful, the model has the potential to be scaled and
replicated in other developing nations facing similar challenges, contributing to a global
improvement in road safety practices.
7. Academic Contribution:
The study contributes to the academic community by advancing the understanding of the
practical application of AI in traffic management and road safety. The research outcomes,
methodologies, and insights can serve as a valuable resource for future studies in the intersection
of AI and transportation.

1.6 Scope and Limitations

1.6.1 Scope

11
The scope of the project is to develop a smart traffic management system that utilizes YOLOv5
and AlexNetV3 Convolutional Neural Network Architecture for the detection of emergency
vehicles and the computation of average vehicle area. The system aims to optimize traffic flow
and improve emergency response times by accurately detecting emergency vehicles and
dynamically adjusting traffic signals to ensure priority access for the emergency vehicle.
1.6.2 Limitations
The limitations of the project include:
1. Hardware Limitations: The proposed system requires a high-resolution camera and a powerful
processor to perform real-time vehicle detection and traffic monitoring. The system's
performance may be limited by the hardware's processing power and memory capacity.
2. Limited Dataset: The accuracy of the system's vehicle detection and emergency vehicle
identification may be limited by the size and quality of the dataset used for training the models. A
larger and more diverse dataset may improve the system's accuracy.
3. Environmental Factors: The system's performance may be affected by environmental factors
such as lighting conditions, weather, and road conditions. These factors may affect the accuracy
of the vehicle detection and emergency vehicle identification.
4. Cost: The proposed system may require significant investment in hardware and software
development, which may limit its adoption in low-income areas or regions with limited resources.
5. Limited Generalization: The proposed system is designed for emergency vehicle detection and
traffic management in urban areas. Its effectiveness in other settings or regions may be limited.

1.7 Thesis Structure

This project is organized into chapters; each introducing new ideas in the whole project report.
Chapter one presents an introductory aspect of the project, the background of road safety and
traffic management systems of The Gambia, the problem statement, the proposed solution to the
problem statement, and contributions. It also presents the motivation, interest, and significance of
the project and the aims and objectives to point out why a new system like this must be
developed and implemented. This chapter also points out the scope and limitations of the project.
Chapter two presents existing research on road safety and traffic management in The Gambia and
worldwide and previous work done on vehicle detection, easy traffic flow, prevention of road
accidents, and emergency vehicle detection. It explains road safety measures, AI, and machine
learning techniques used in traffic management systems. It also entails a comparison of different
approaches to traffic management. Chapter three deals with AI and machine learning
technologies, models used in the project, methodology, and detailed analysis of the methodologic
used to bring up the system. It also explains the Road Safety and Traffic Management system of
the Gambia, the feasibility study done, AI and machine learning models used to enhance the
proposed traffic management system, and the limitations of the techniques and models of the
12
project. Chapter four deals with a comparative analysis of typical methods used in solving the
automatic traffic light control problem, experimenting with multiple data sets, and analyzing their
respective strengths and weaknesses. It also explains the use case diagrams. Chapter five
discusses the hardware and software architectures of the system and operational flow charts,
system requirements, and general requirements. It also explains the programming languages,
development environment, and technologies used. It further describes the proposed system
architecture, TMS architecture, and interface design. Chapter Six delves into traffic control
systems based on machine learning, data processing, image processing, pros and cons of the
system. Chapter seven includes the implementation of the smart traffic management system,
testing, scope of testing, input data, and validation of the system. Chapter eight includes a
conclusion, references, and appendices.

Reference

[1] Orosz, Gabor & Krauskopt, B & Wilson, RE. (2006). Traffic jam dynamics in a car following model
with reaction-time delay and stochasticity of others. IFAC Proceedings Volumes (IFAC-PapersOnline). 6.

[3] Njie, Lamin. (2023, January 5). Gambia Losses Over Two Billion Dalasis Due to Road
Traffic Crashes. The Voice Gambia Newspaper.
https://www.voicegambia.com/2023/01/05/gambia-losses-over-two-billion-dalasis-due-to-road-
traffic-
crashes/#:~:text=%E2%80%9CBased%20on%20the%20statistics%20obtained,traffic%20acciden
ts%20in%20the%20Gambia.

[4] Kandeh, A. M., Ceesay, A., & Colley, E. (2020). Exploring the Causes of Road Traffic
Accidents in Urban Gambia from 2014 to 2018. Global Scientific Journal, 8(10), 2244-2265.

[5] Chava, V., Nalluri, S. S., Kommuri, S. H. V., & Vishnubhatla, A. (2023). Smart Traffic
Management System using YOLOv4 and MobileNetV2 Convolutional Neural Network
Architecture. Proceedings of the Second International Conference on Applied Artificial
Intelligence and Computing (ICAAIC 2023). IEEE Xplore Part Number: CFP23BC3-ART;
ISBN: 978-1-6654-5630-2.

[6] World Health Organization. (2015). Road safety in the African Region, 2015. Retrieved from
https://www.who.int/violence_injury_prevention/road_safety_status/2015/

13