Duraimurugan.N1, Dr. S. Gnanavel2, Abirami.C3, Anto Jevin A V4, Aravindan.

S5

Turkish Online Journal of Qualitative Inquiry (TOJQI)

Volume 12, Issue 3, June 2021: 2221-2228

Research Article

Clearing Pathway for Ambulance using YOLO and Raspberry Pi

Duraimurugan.N1, Dr. S. Gnanavel2, Abirami.C3, Anto Jevin A V4, Aravindan.S5

Abstract
In recent times, traffic congestion has been one of the most concerning factors. It not only
affects our scheduled lives but also brings about a major threat to patients in need. Vehicle
blockage could be a factor deciding the fate of one’s life. The goal of the project is to help the
ambulance reach the nearest hospital in the least possible time quickly and safely. In order to
achieve this traffic signals are being monitored and controlled, using microphone which
distinguishes sound decibel level of ambulances to that of other vehicles. In addition to this, a
camera takes pictures of jammed traffic signals for further verification. On successful detection,
the signal will be made to turn green, hence clearing a pathway for ambulance.
Keywords—Internet of things (IoT), Raspberry Pi, YOLO Algorithm, Mobile Notification

INTRODUCTION
I.
Today’s modern world requires fast processing of almost everything. With this busy life
accidents occur from time to time and place to place. Ambulance is the only emergency vehicle that
saves the affected one’s life. There are people who do not care much to clear a path for the
ambulance to move further. In this regard, we create a model where ambulances are detected from
long distances in a traffic signal or a junction and clear path for them, so that they could move
towards their desired destination without any delay. The main objective of this system is to help
emergency vehicles reach their destination in time. Traffic congestions are the primary reason for
ambulances not reaching their destination in time. To rectify this issue, the project proposes a
system that will help clear traffic immediately when an emergency vehicle is detected.
_________________________________________________
1,2
Assistant Professor, Department of CSE, Rajalakshmi Engineering College, Chennai - 602 105, India
3,4,5
Student, Department of CSE, Rajalakshmi Engineering College, Chennai-602 105, India

II. TECHNOLOGIES
Raspberry Pi
A.
Raspberry Pi is a progression of little single-board PCs created in the United Kingdom by the
Raspberry Pi Foundation in relationship with Broadcom. The Raspberry Pi project initially inclined
towards the advancement of showing fundamental software engineering in schools and in creating
countries. The first model turned out to be more famous than anticipated, selling outside its
2221
 Clearing Pathway for Ambulance using YOLO and Raspberry Pi

objective market for utilizations like mechanical technology. It is broadly utilized in numerous
regions, for example, for climate monitoring, in view of its minimal effort, seclusion, and open plan.
It is normally utilized by PC and electronic specialists, because of its appropriation of HDMI and
USB gadgets. There exist a 5 Generations in the Raspberry Pi family which are the Raspberry Pi,
Raspberry Pi 2, Raspberry Pi Zero, Raspberry Pi 3, Raspberry Pi 4, Raspberry Pi Pico.

B. Machine Learning
Machine learning model predictions allow businesses to make highly accurate guesses as to the
likely outcomes of a question based on historical data, which can be about all kinds of things –
customer churn likelihood, possible fraudulent activity, and more. Both machine learning and
predictive analytics are used to make predictions on a set of data about the future. Predictive
analytics uses predictive modeling, which can include machine learning. Predictive analytics has a
very specific purpose: to use historical data to predict the likelihood of a future outcome

C. Raspberry Pi Camera
The camera board appends to the Raspberry Pi by means of a 15-way lace link. There are just
two associations with make: the lace link should be appended to the camera PCB, and to the
Raspberry Pi itself. You need2 to get the link the correct route round, or the camera won't work.
The Best preferred Raspberry Pi Camera of 2021 is the AUKEY PC-LM1 1080p Webcam. The
Raspberry Pi High Quality Camera addresses a jump forward in quality, clearness and definition.
It additionally opens up conceivable outcomes to utilize proficient DSLR focal points, for example,
the Canon EOS EF line, I found a connector for around 50 GBP, in the event that you get one,
discover one with gap control.
D.Raspberry Pi Microphone
USB Microphones are the least demanding method of getting an amplifier working with your
Raspberry Pi. Perhaps the main benefits of utilizing a USB amplifier is that it is fitting and play.
The Raspbian working framework will naturally recognize the receiver when it’s connected.
E. OpenCV
OpenCV (Open Source Computer Vision Library) is a library of programming capacities for the
most part focused on on going PC vision. Originally created by Intel, it was subsequently upheld
by Willow Garage then Itseez (which was subsequently procured by Intel). The library is cross-
stage and free for use under the open-source Apache 2 License. Beginning with 2011, OpenCV
highlights GPU speed increase for continuous activities.
F. YOLO Algorithm
YOLO (You Only Look Once) is a viable on going item acknowledgment calculation, first
portrayed in the fundamental 2015 paper by Joseph Redmon et al. Picture grouping is one of the
many energizing uses of convolutional neural organizations. Beside straightforward picture
arrangement, there are a lot of captivating issues in PC vision, with object recognition being perhaps
the most intriguing. It is ordinarily connected with self-driving vehicles where frameworks mix PC
vision, LIDAR and different innovations to produce a multidimensional portrayal of the street with
every one of its members. Item location is additionally ordinarily utilized in video reconnaissance,
particularly in swarm observing to forestall fear based oppressor assaults, tally individuals for
general insights or examine client experience with strolling ways inside malls.

2222
 Duraimurugan.N1, Dr. S. Gnanavel2, Abirami.C3, Anto Jevin A V4, Aravindan.S5

EXISTING SYSTEM
III.
Sensors were used for measurement of traffic density using RFID technique. The ambulance
driver carries an android phone with the software installed in it and sends a message to traffic signal
controller at the traffic junction.
A system which finds the quickest path by controlling traffic light signals in favour of
ambulance. The time delay is reduced by applying the RF technology that controls the traffic signals
System uses sound detecting instruments that are trained with various vehicles’ sounds. On
detection of an emergency vehicle, the traffic controller is alerted and is asked to change the traffic
signal accordingly

A. DisadvantagesOf Existing System

• Cannot be used to identify the density of the traffic beyond a particular distance [8].
• Ambulance driver has to take his hand held device to send notification. Requires presence of
a human to change signals [9].
• Requires presence of a human to change signals [6].
• System requires to be recharged after every 4 hours [7].
• Sound of siren detected is not necessarily that of an ambulance.
IV. LITERATURE SURVEY
This project uses ultrasonic sensors to detect traffic in roads and intimates the drivers to take
alternative routes. This system can detect traffic of up to 3 kilometres [1].
The system detects the presence of an ambulance by observing the sound pattern of the traffic.
The sound of a siren would alert the system to take necessary actions [2].
This system uses a GPS module that keeps track of the location of the ambulance. The system
alerts the ambulance driver about the traffic ahead of him by comparing his location to the live
traffic feeds of the city [3].
This system uses sound detecting instruments that are trained with various vehicles’ sounds. On
detection of an emergency vehicle, the traffic controller is alerted and is asked to change the traffic
signal accordingly [4].
This system uses a combination of image processing and RFID. The image detects the presence
of a vehicle and intimates the traffic controller. The emergency vehicle driver has to use his RFID
to pass the information to the next signal that he is approaching them [5].
This article summarizes the various types of telemedicine and mobile health (mHealth) tools and
their practical applications and benefits for patient care. Artificial Intelligence pursues building the
computers or machines as brainy as human beings [6], FastConvolutional Neural Network the
algorithm will not look at the image completely but in YOLO the algorithm looks the image
completely by predicting the bounding boxes using convolutional network and the class
probabilities for these boxes and detects the image faster as compared to other algorithms [7]. The
system uses RNN and CNN network for action recognition either in the form of images or signals.
Combining CNN and RNN will enhance the ability to recognize different actions at varied time

2223
 Clearing Pathway for Ambulance using YOLO and Raspberry Pi

span [8], The system uses RNN and CNN network for action recognition either in the form of
images or signals. Combining CNN and RNN will enhance the ability to recognize different actions
at varied time span [9].
V. PROPOSED SYSTEM
The proposed system automatically controls traffic signals when the presence of an ambulance
is detected. The system receives data in the form of sound and image. It is trained to recognize if
the sound received is of an ambulance’s siren or not. The system double checks this through visual
feed where pictures of the ambulance are taken and checked using image processing. On successful
detection and confirmation, the system immediately turns the signal green in the direction in which
the ambulance is travelling, leaving all the other signals red.
A. Proposed
System Advantages
• Double verification to check the presence of an ambulance.
• Intimates the other drivers to give way for the emergency vehicle immediately.
• Does not require any physical presence to control signals.
• Driver and police men get instant notifications on their mobile phones.
VI. MODULES
The project “Clearing Pathway for Ambulance” aims at providing added functionalities to the
already existing traffic signal systems. The addition of this system will enable the traffic signals to
change signals automatically and help the emergency vehicle cross the signal without having to
wait in long and congested roads.
A. Detection of Ambulance
This module contains a microphone and a camera that is on alert mode. The microphone picks
up live traffic sound feeds and monitors the decibel level. If the decibel level matches an
ambulance’s siren, then the Raspberry Pi system is alerted for further actions. The camera picks up
live traffic feeds and looks for a potential presence of an ambulance. If confirmed, a signal is sent
to the system which will alter the traffic signals accordingly.
B. Traffic Light Control
This module controls the 4 way traffic signal. The traffic lights of all the four lanes are controlled
with specific sleep timers. Each signal transition occurs with a pre-defined buffer time to avoid
vehicles over speeding to try cross the signal before the lights turn red. The signals turn green on
the lane the ambulance is travelling when the system successfully detects an emergency vehicle.
C. Mobile Notification
This module consists of a mobile application that traffic policemen and ambulance driver could
use. This app is designed with the intention of passing information related to the particular traffic
signal to the user. The application delivers a notification to the user when an ambulance is detected
and that the signal has turned green. This is done to ensure that the traffic police on duty is aware
of the situation and is able to change signals manually if the signal in not under automatic transition
mode.

2224
Duraimurugan.N1, Dr. S. Gnanavel2, Abirami.C3, Anto Jevin A V4, Aravindan.S5

VII. ARCHITECTURE DIAGRAM

Fig: 1.1 Architecture Diagram

VIII. WORKFLOW DIAGRAM

Fig: 1.2 Workflow Diagram

2225
 Clearing Pathway for Ambulance using YOLO and Raspberry Pi

IX. MODEL ARCHITECTURE

Fig: 1.3 Model Architecture

X. EXECUTION IMAGES
This section contains some of the execution images for better understanig for the end uses
A. Detection of ambulance

B. Traffic Light Control

2226
 Duraimurugan.N1, Dr. S. Gnanavel2, Abirami.C3, Anto Jevin A V4, Aravindan.S5

C. Mobile Notification

XI. CONCLUSION
Traffic jam is one of the most common problems faced in our day to day lives. It plays a crucial
part of saving people’s life in one way or the other. In our country, emergency vehicles like
ambulances travel in the same track way as the public do. Traffic congestions in these routes are
directly or indirectly responsible for the life of the patients travelling in the emergency vehicle to a
greater extent. Ambulances are lifesaving vehicles that should save life in minimal time. Clearing
pathway for ambulance is a system developed in order to give a solution to this problem. In this
system, siren sound of emergency vehicles is detected from up to 250m and images of ambulance
are verified with the help of camera installed in the traffic lights. Hence, the road in which the
ambulance travels is cleared by making signals automatically switch to green, ensuring that the
emergency vehicles cross the signal without any delay. In future, the above system can be enhanced
by updating the android app that can serve the ambulance driver with the nearest hospital location
and also if that particular hospital can treat the patient who is travelling in that ambulance or not.
The system can also be trained to achieve better accuracy and handle situations with multiple
ambulances
REFERENCES
[1] Roopashree V, Nikitha Bai E, Malavika D N, Suman A, “Traffic Congestion Detection and
Alerting”, IJERT, July 2020, vol 09, Issue 07, page no 1339
[2] Reeta R, Kirithiga R, Kavitha V. Kumar, Jaishree M, “IoT- Based Traffic Signal Control for
Ambulance”, IJEAT, Feb 2020, vol 09, Issue 03, page no 3250
[3] Usha NS, Aritha S, Viswathika S, “Make way for ambulance using IoT”, IJIRSET, March
2018, vol 7, Issue 2, page no 142
[4] Reeta R, Krithiga R, Kavitha V Kumar, Jaishree M, “IoT based Traffic Signal Control for
Ambulance”, IJPAM, January 2018, vol 9, Issue 3, page no 3250
[5] Ashit S Chitta, Dinesha P, “Priority Management of Emergency Vehicles using IoT
Approach”, IJARCCE, September 2016, vol 5, Issue 9, page no 465
[6] Duraimurugan, N. Manoj Kumar, B. Malini, C. Kowsalya, R., “Learnart: Drawing
environment using convolutional neural networks”, International Journal of Recent
Technology and Engineering, 2019, 8(2 Special Issue 3), pp. 770–772

2227
 Clearing Pathway for Ambulance using YOLO and Raspberry Pi

[7] Geethapriya. S, N. Duraimurugan, S.P. Chokkalingam, “Real time object detection with yolo”,
International Journal of Engineering and Advanced Technology (IJEAT) ISSN: 2249 – 8958,
Volume-8, Issue-3S, February 2019
[8] Nandhini, R. Duraimurugan, N. Chokkalingam, S.P, “Face recognition based attendance
system”, International Journal of Engineering and Advanced Technology, 2019, , 8(3 Special
Issue), pp. 578–581
[9] J.Rajalakshmi, N.Duraimurugan, S.P.Chokkalingam, “Action recognition for controlling
electronic appliances” International Journal of Engineering and Advanced Technology, 2019,
8(3 Special Issue), pp. 565–567
[10] F. Meucci, L. Pierucci, E. Del Re, L. Lastrucci, P. Desii, “A Real-Time Siren Detector To
Improve Safety Of Guide In Traffic”, EUSIPCO, August 2018, vol 16, Issue 2, page no 1221
[11] Karthika K, Lavanya G, Pavithra M, Periyathambi P, “Advanced Ambulance Monitoring
System using IoT”, IRJET, February 2020, vol 7, Issue 2, page no 479
[12] Karthik B V, Manoj M, Rohit R Kowshik, Akash Aithal, Dr. S. Kuzhalvai Mozhi, “Ambulance
Detection and Traffic Control System”, IRJET, April 2019, vol 6, Issue 4, page no 1128
[13] Venkatesh H, Shrivatsa D Perur, Jagadish M C, “An Approach to Make Way for Intelligent
Ambulance Using IoT”, IJEER, March 2015, vol 3, Issue 1, page no 218

2228