Download as pdf or txt
Download as pdf or txt
You are on page 1of 9

Project Synopsis

On

Vehicle Recognition and Traffic Tracking Surveillance System

Submitted By
Yash Hulke (C-60)
Vijendra Mishra (C-58)
Purva Baniya (B-26)

Under the guidance of


Neha Purohit
Department of CSE

Department of Computer Science & Engineering


G.H. Raisoni College of Engineering
(Autonomous Institute Affiliated to Rashtrasant Tukdoji Maharaj Nagpur University, Nagpur)
Near CRPF gate, Digdoh Hills, Nagpur

Session 2020-2021
INDEX

● ABSTRACT

● INTRODUCTION

● OBJECTIVES

● LITERATURE SURVEY/STUDY OF EXISTING

SOLUTIONS/PRODUCTS

● PROPOSED METHODOLOGY/ SYSTEM ARCHITECTURE

● HARDWARE / SOFTWARE SPECIFICATION

● CONCLUSION

● REFERENCES
ABSTRACT

The rapid growth of computer vision techniques has turned video-based traffic monitoring into an
excellent alternative of traditional monitoring measures. At present, vehicle detection, tracking and
speed estimation are three main tasks for most road monitoring systems. However, systems that
integrate all these functionalities are scarce. Moving vehicle detection, tracking, and counting are
very critical for traffic flow monitoring, planning, and controlling. Video-based solution, compared to
other techniques, does not disturb traffic flow and is easily installed. By analyzing the traffic video
recorded from a video camera and applying blob detection methods along with other OpenCV
development kits ,we will be able to detect, track, count moving vehicles accurately and predict the
speed of moving vehicles.
INTRODUCTION

Nowadays, with the growth of motorization and urbanization, traffic congestion and traffic violations
have become two urgent problems in transport and traffic management as they increase the
burden of governments and local authorities to a great extent. Generally speaking, a good solution
for solving these problems is to emphasize infrastructure construction, which may enhance the
traffic capability for road networks. However, with the restriction of limited space ,money and other
resources, people are now focusing on improving the operational efficiency by employing various
high-technology systems, like Intelligent Transportation System (ITS).

One such system keeping up with the advancing video capturing technology is video surveillance.
Nowadays, video surveillance systems are also being deployed at department stores in order to
provide advertising assessment and quality of service, on highways for traffic monitoring purposes,
and even on houses for elderly people to assist them in a non-invasive manner.
The same non-invasive methodology can be used for better management and monitoring of traffic
while also being more cost efficient compared to existing systems.

While the existing systems like Magnetic Loops, Pressure tubes , Radar Guns can provide
analysis of some of the features of flowing traffic, a video based monitoring system can provide all
the provided functionalities and more.

Using the video footage obtained from video recording cameras placed strategically to monitor
traffic flow many components of the vehicles passing through in the footage can be determined
using the VRTTS system.

Some of the features identified are :

• Compare with other methods, stationary camera has a more extensive field of vision.

• Low cost on installation and maintenance, installation and maintenance without transportation
interruption.

• Obtain traffic information, include traffic volume, vehicle speed synchronously.


OBJECTIVES

The aim of this project is to develop a Video-based Road Traffic Surveillance System which is able
to detect vehicles, gather traffic information and predict speed by analysing the traffic video.
Objectives of our project are as follows:

1. Identify a robust background generation method that is able to produce high quality
background images by overcoming the interference of illumination change and
stationary objects.

2. Improve the background subtraction algorithm in order to obtain more complete


vehicle blobs and overcome blobs overlapping problems.

3. Algorithm to track the vehicle blobs and obtain the count and trajectories of vehicles.

4. speed estimation model to estimate vehicles’ speed with help of blob detection and
tracking.
LITERATURE SURVEY / STUDY OF EXISTING SOLUTIONS

While there exists systems that can

A tabulation of the information that can be collected and the Advantages-Disadvantages of the
existing systems is illustrated below :

Monitoring Information that can Advantages Disadvantages


System be collected
Computer vision I.Vehicle detection ● Many information ● Not reliable in
2.Count can be collected varying lighting condition
3.Speed ● More than one lane and
4.Category can be monitored weather
5.Path ● Easy to install ● Field of view must
6.Flow rate ● Low maintenance be reasonably clear, free
7.Queue length & ● Autonomous from occlusions
Route travel times ● Flexibility in ● Vehicles must be
9.Lane changes changing the traffic scene separated from each other
● Information can be so they do not appear
relayed back to connected
authorities in real-time
● Easy to be
networked
Magnetic loop I.Vehicle detection ● Reliable ● Dig up the road
2.Count ● Low maintenance ● Costly to install
● Autonomous ● Limited
information
Pressure tubes 1.Vehicle detection ● Reliable ● Limited
2.Count information
Radar gun l.Speed ● Accurate speed ● Limited
measurement information
Microwave I.Vehicle detection ● Reliable ● Costly to install
sensors 2.Count ● Limited
information
PROPOSED METHODOLOGY/ SYSTEM ARCHITECTURE

VRTTSS is composed of four modules:

1. Vehicle detection

2. Vehicle tracking

3. Vehicle counting

4. Speed estimation.

The Vehicle detection component is responsible for segmenting any foreground


objects of interest from the background.
The Vehicle tracking component is to match the same image blob appeared in a
sequence of frames.
The Vehicle counting component is to keep count of the blobs detected.

The speed estimation component is to predict the speed of the vehicle.


HARDWARE / SOFTWARE SPECIFICATION

Software

o Development toolkit [Microsoft Visual Studio 2015]/Python

o OpenCV 3.0 and above

o Windows 7 and above

Hardware

o Processor intel i3 and above

o RAM 4GB and above

o Storage 10GB and above


CONCLUSION

Implementation of traffic monitoring system techniques , which include background


subtraction, object tracking and vehicle recognition for vehicle count and speed
estimation
This approach can then be used in applications which require real time traffic
surveillance systems with additional features such as lane checking , classification of
vehicles can be implemented with further study

REFERENCES

1. Davaadorj Norpel, Sodkhuu Dalaikhuu, Khurelbaatar Tseveenjav. Traffic

Surveillance System based on Computer Vision and Its application , Institute

of Electrical and Electronics Engineers [IEEE] , 2015

2. http://opencv.org/

You might also like