Scribd
Upload
31 views13 pages

Project Presentation Format

This document presents a project proposal for a traffic management system for ambulances using machine learning. The system aims to reduce loss of lives during emergencies by clearing traffic for ambulances more efficiently. It works by using video input, detecting vehicles including ambulances using deep learning models, and then controlling traffic signals based on ambulance locations to provide clearer paths. The methodology involves phases of vehicle detection and counting, ambulance detection, and taking traffic control actions based on the data. The system has been tested with a raspberry pi, camera, and GPS module. It has also been trained on a dataset of over 5000 images to achieve 98.5% accuracy in detecting ambulances.

Uploaded by

Richa Mishra
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
31 views13 pages

Project Presentation Format

This document presents a project proposal for a traffic management system for ambulances using machine learning. The system aims to reduce loss of lives during emergencies by clearing traffic for ambulances more efficiently. It works by using video input, detecting vehicles including ambulances using deep learning models, and then controlling traffic signals based on ambulance locations to provide clearer paths. The methodology involves phases of vehicle detection and counting, ambulance detection, and taking traffic control actions based on the data. The system has been tested with a raspberry pi, camera, and GPS module. It has also been trained on a dataset of over 5000 images to achieve 98.5% accuracy in detecting ambulances.

Uploaded by

Richa Mishra
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 13

A Project on

“TRAFFIC MANAGEMENT SYSTEM FOR AMBULANCE USING ML”


Vision-Aid
by
1. Richa Mishra (0863AB201018)
2. Rudra Mandloi (0863EC201027)
3. Ujjwal Singh (0863EC201034)

Under the guidance of


Dr. Mahendra Thakur
&
Prof. Rajendra Shimpi

PRESTIGE INSTITUTE OF ENGINEERING MANAGEMENT & RESEARCH,


INDORE

DEPARTMENT OF AUTOMATION & ROBOTICS


DEPARTMENT OF ELECTRONICS & COMMUNICATIONS
1
Problem Statement/Domain

• Reduction in loss of lives during emergency cases due to lack of


proper traffic management systems.
• This Project of ours falls under the domain of
MedTech/HealthTech.

2
Introduction and Background
• One of the most profound aftermaths of evolving technologies in this
modern era is rapidly increasing vehicular counts.
• Traffic congestion has become a serious problem in most countries around
the world.
• Moreover, the rate of road expansion is just one-third the vehicular growth
rate in countries like India, which further worsens the problem.
• Due to increased traffic, there has been an increase in traffic jams, which
has caused a problem for ambulances in moving patients to the appropriate
destinations on time in case of emergency.
• Statistically, it has been observed that more than 20% patients requiring
urgent medical attention die on their way to the hospital because of
delays.

3
WORKING

• Input Source(camera): This is the source of the input data used for ambulance detection.
It could be a live video feed from a camera or a recorded video.
• Video Preprocessing: The input video data often requires preprocessing to enhance the
quality and extract relevant information. This stage may involve operations such as noise
reduction, image stabilization, and image enhancement
• Object Detection: In this stage, an object detection algorithm is applied to identify
potential objects in the video frames for this approach object detection is using deep
learning-based models s YOLO used. The algorithm scans the frames and generates
bounding boxes around potential objects.
• Ambulance Verification: In some cases, additional verification steps are performed to
reduce false positives. For example, the system may check for specific features or patterns
that are unique to ambulances, such as flashing lights, siren sounds, or distinct color
patterns. This stage helps increase the accuracy of ambulance detection.
• Output Display/Action: The final stage involves displaying the output to the user or taking
appropriate actions based on the detection results. activating audible alarms, or
triggering automated response systems.

4
Methodology
• THE PROPOSED SYSTEM AND HARDWARE USED
• Descriptions of the hardware components are provided below. In addition to the
above-mentioned, the system also includes subcomponent named Traffic Control
Room. The traffic control room will help the ambulance to reach its destination on
time by clearing the route for the ambulance and by changing the traffic signals
whenever and wherever needed.
• HARDWARE COMPONENTS USED
• 1. Raspberry Pi
• 2. Camera
• 3. GPS Module

5
BLOCK DIAGRAM

6
METHODOLOGY
• Our project consists of three phases:

1. Vehicle Detection and Counting

2. Ambulance Detection

3. Decisions Based on Data

• Vehicle Detection and Counting Following are the steps involved in this phase:

1. Camera Positioning

2. Image Subtraction

3. Vehicle Counting

4. Data Processing

7
METHODOLOGY
• Machine Learning Algorithms:
Convolutional Neural Networks (CNNs), Support Vector Machines (SVMs),
and Random Forests.

• These algorithms are used to perform tasks such as object recognition,


image processing, and data analysis.

8
Current Progress
• Made and Tuned Traffic Light part using NodeMCU.
• Currently testing it with RFID for variety of uses.
• Trained Dataset of around 5000 images.
• Craft work of model is partially done.

9
Current Progress
Training Graph of Dataset showing acquired accuracy as of now by
rigorous training. (98.5%)

10
Machine Learning Image Training Using Darknet

11
References
1. V. Srinivasan, Y. Priyadharshini Rajesh, S. Yuvaraj, and
• M. Manigandan, “Smart traffic control with ambulance detection,” IOP Conf.
Ser. Mater. Sci. Eng., vol. 402, no. 1, 2018.
1. Nation, “20 per cent of emergency patient deaths blamed on traffic jam delays,”
Nation Thailand, 2017. [Online]. Available: https://
www.nationthailand.com/national/30304268. [Accessed: 29-Jan-2020].

2. IRF, “IRF Annual Report,” 2018.

3. PTI, “Delay In Reaching Hospital Caused Newborn’s Death, Says Probe


Report,” NDTV, 2018. .
4. R. Sunitha Rao, “When city traffic halts ambulances, lives,” Times of India,
2014. .
5. C. Morella and J. Melvin, “Patients die as Manila traffic jams block
ambulances,” ABS-CBN News, 2019. [Online]. Available: https://news.abs-
cbn.com/news/09/09/19/patients-die-as-manila-traffic- jams-block-ambulances.
[Accessed: 29-Jan-2020].
12
Thank You

13

You might also like