0% found this document useful (0 votes)
122 views12 pages

Smart Traffic Management System Using Machine Learning

The document describes a Mask R-CNN model for vehicle detection in traffic management. The model detects vehicles in images and video from traffic cameras to count vehicles and allocate green light times. It was trained on 131 images and tested on 21 images, achieving 98.9% accuracy. The Mask R-CNN model outperformed other R-CNN models like Fast R-CNN and Faster R-CNN for this task based on metrics like accuracy, loss, and processing time. The system will help optimize traffic flow and reduce waiting times.

Uploaded by

Vedant Agrawal
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
0% found this document useful (0 votes)
122 views12 pages

Smart Traffic Management System Using Machine Learning

The document describes a Mask R-CNN model for vehicle detection in traffic management. The model detects vehicles in images and video from traffic cameras to count vehicles and allocate green light times. It was trained on 131 images and tested on 21 images, achieving 98.9% accuracy. The Mask R-CNN model outperformed other R-CNN models like Fast R-CNN and Faster R-CNN for this task based on metrics like accuracy, loss, and processing time. The system will help optimize traffic flow and reduce waiting times.

Uploaded by

Vedant Agrawal
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/ 12

Paper ID: 143

Mask R-CNN Based Vehicle Detection Model for Traffic


Management System

FIRST
INTERNATIONAL CONFERENCE ON INNOVATIVE
APPLICATIONS OF EMERGING TECHNOLOGIES AND
MANAGEMENT,(ICIAETM)-2021
Authors:
Vedant Agrawal, PIEMR, Indore
Sushmita Kaviraj, PIEMR, Indore
Diksha Pal, PIEMR, Indore
Prakhar Verma, PIEMR, Indore
Abstract
In the current era of technology every continuous or discontinuous work is
going to be converted into automation. We all knew that the next revolution is
industry 4.0 which is going to replace all such types of activities. On a bigger
note, automation is going to achieve by replacing human resources i.e. 5 major
senses of human sight, sound, smell, touch, and taste, and all depends upon
the type of work. The prescribed project also works with one of the senses for
automation ie VISION. We developed a device that detects the object and
works accordingly. The main objective of the paper is to provide a system that
works efficiently with industries to work faster and effectively. In our paper,
we have focused comparative study on vehicle detection model which can
detect the vehicle and also show the accuracy percentage on the block, for
that, we collect data, then pre-process the data, model building, and most
importantly, compare models and evaluate to get better results.
Content
• Introduction
• Objectives
• Methodology
• Results and Discussion
• Conclusion
Introduction
• The Project is entitled as a Mask R CNN based vehicle detection
model for traffic management system. It is a working application of
vehicle detection. The model replaces the heavy implementation of
electronic sensors in terms of cost and maintenance by using pre-
installed cameras at the signals.
• The system is fully automated and also able to detect the number
plates if trained the model for that too, which helps in reducing
traffic violation by challan or a voice note on speaker for such, and
also helps to detect crime vehicle easily as the system
automatically detects the vehicle and also provides the list of
number plates make easy to track.
Objectives
• The main objective is to train a machine learning model which is
used to detect the vehicle and count. it using the number of
counter box draw around it, this will help in the time allocation
algorithm and pass more time to heavily trafficked lanes in
comparison to other lanes.

• The project results in some crucial or indirect savings like it


reduces the time for a vehicle to wait which leads to saving in fuel,
time, environment, and money.
Methodology
• The Proposed system work with both video and images, The system
first takes the video input from all four of the lanes (here we
assumed that we have 4 lanes in comparison) then divide them into
timestamps, the parallel system detects the vehicle using the input
video and count each one of the. It updates the count after every 10
sec to get more accurate results.

• We also include a deadlock prevention condition and a parallel


comparison algorithm for lanes having the minimum number of
vehicles than the other three always and if two or three of lanes
having the same number of vehicles respectively
Methodology
Methodology
• The process flow inside the model can be described as At each time
instance (after 10 sec) a image frame is captured stated as input
frame now checking of the lane number i.e. the input frame is from
which lane after which the input frame is processed under Mask
RCNN algorithm by finding Region Of Interest (ROI) and detecting
vehicle simultaneously for making mask it proceed by extracting
new corner points and assigning them to detected segment then
tracking old and new corner points and by deleting points out of ROI
then proceed to final vehicle counting.
Results and Discussion

The Loss V/S Epoch and Accuracy


V/S Epoch graph is shown above of
the trained model using the
proposed system algorithm i.e. Mask
RCNN. The model is trained using
131 training and 21 test images and
evaluated against a different dataset
Results and Discussion

The table is the result


comparison of the
different algorithms.

The results are


compared by other 3
results on various
parameters like test
time per image, mAP
accuracy, the loss
against similar
dataset provided as
coco
Conclusion
On a conclusive note, we can say that there are many ways to train a
convolution neural network model, along with making some aspects
constant like dataset and other training parameters then to it all differs
because of the algorithm used to train the model. On other hand, the
comparative study of algorithms of regional convolutional neural
network or RCNN comes out as in between RCNN, Fast RCNN, Faster
RCNN, Mask RCNN the Mask RCNN comes out the best way among
all of them with an accuracy of 98.9%.
Thank You!

You might also like