Real time lane detection system using Python and OpenCV

Major/Minor Project Report submitted in the partial

fulfillment of the requirements for the degree of

BACHELOR OF TECHNOLOGY
in
<Computer Science>

By

<<Tanuj Raghav, Arth Tyagi, Tushar Malik, Rakshit Verma>>

<< >>

Under the Supervision of

<<Vijayalaxami >>
(with designation & Deptt)

SCHOOL OF ENGINEERING AND TECHNOLOGY

2018-2022
i
 REPORT APPROVAL

This project report entitled (Title) by (Author Name) is approved for the
degree of ___________ (Degree details).

Examiners
……………………………..

Supervisor
……………………………..

Date: -------------------

Place: ------------------
ii
 DECLARATION

I declare that this report represents my ideas in my own words and where others' ideas or
words have been included, I have adequately cited and referenced the original sources. I also
declare that I have adhered to all principles of academic honesty and integrity and have not
misrepresented or fabricated or falsified any idea/data/fact/source in my submission. I
understand that any violation of the above will be cause for disciplinary action by the
University and can also evoke penal action from the sources which have thus not been
properly cited or from whom proper permission has not been taken when needed.

Signature of the student ……………………..

Name of Student ……………………..
Date ……………………..

iii
 TABLE OF CONTENTS

ACKNOWLEDGMENTS i
REPORT APPROVAL ii
DECLARATION iii
ABSTRACT iv

1. Introduction 1
1.1 Overview
1.2 Objective and scope

2. Literature survey __
2.1. Previous work
2.2. Problem Identification/Motivation

3. Methodology

4. Steps / Source Code __

5. Results/Implementation/Applications __

6. Conclusions
Conclusion/Discussion/Suggestions for Further Study/Improvement for
future work

References __
...........................................................................................................
..............................................................................................................
..............................................................................................................
..............................................................................................................
............................................................................................... __
 ABSTRACT

Driver assistance system is a technology used to make motor vehicle travel safer
by automating, improving or adapting some or all of the tasks involved in
operating a vehicle. Driver assistance serves to make travel comfortable and
easier, while also increasing car and road safety. While some systems help with
the task of driving, others alert the driver to errors or hazards, such as lane
departure detection and drowsiness detection. Aside from vehicle control, driver
assistance can also refer to secondary driving tasks such as location finding,
route planning and obstacle detection. Driver assistance is a developing field.
This paper aims at detecting lanes using Python and OpenCV. In real time
vehicular movements will be captured using a camera and the same will be
processed to achieve the goal. The Hough Transform is used to detect lanes in
an image or video.
In this research work, we explore the vehicle
detection technique that can be used for traffic surveillance
systems. This system works with the integration of CCTV cameras
for detecting the cars. Initial step will always be car object
detection. Haar Cascades are used for detection of car in the
footage. Viola Jones Algorithm is used in training these cascade
classifiers. We modify it to find unique objects in the video, by
tracking each car in a selected region of interest. This is one of the
fastest methods to correctly identify, track and count a car object
with accuracy up to 78 percent.
In this research work, we explore the vehicle
detection technique that can be used for traffic surveillance
systems. This system works with the integration of CCTV cameras
for detecting the cars. Initial step will always be car object
detection. Haar Cascades are used for detection of car in the
footage. Viola Jones Algorithm is used in training these cascade
classifiers. We modify it to find unique objects in the video, by
tracking each car in a selected region of interest. This is one of the
fastest methods to correctly identify, track and count a car object
with accuracy up to 78 percent.
In this research work, we explore the vehicle
detection technique that can be used for traffic surveillance
systems. This system works with the integration of CCTV cameras
for detecting the cars. Initial step will always be car object
detection. Haar Cascades are used for detection of car in the
footage. Viola Jones Algorithm is used in training these cascade
classifiers. We modify it to find unique objects in the video, by
tracking each car in a selected region of interest. This is one of the
fastest methods to correctly identify, track and count a car object
with accuracy up to 78 percent.
In this research work, we explore the vehicle
detection technique that can be used for traffic surveillance
systems. This system works with the integration of CCTV cameras
for detecting the cars. Initial step will always be car object
detection. Haar Cascades are used for detection of car in the
footage. Viola Jones Algorithm is used in training these cascade
classifiers. We modify it to find unique objects in the video, by
tracking each car in a selected region of interest. This is one of the
fastest methods to correctly identify, track and count a car object
with accuracy up to 78 percent.
In this research work, we explore the vehicle
detection technique that can be used for traffic surveillance
systems. This system works with the integration of CCTV cameras
for detecting the cars. Initial step will always be car object
detection. Haar Cascades are used for detection of car in the
footage. Viola Jones Algorithm is used in training these cascade
classifiers. We modify it to find unique objects in the video, by
tracking each car in a selected region of interest. This is one of the
fastest methods to correctly identify, track and count a car object
with accuracy up to 78 percent.
In this research work, we explore the vehicle
detection technique that can be used for traffic surveillance
systems. This system works with the integration of CCTV cameras
for detecting the cars. Initial step will always be car object
detection. Haar Cascades are used for detection of car in the
footage. Viola Jones Algorithm is used in training these cascade
classifiers. We modify it to find unique objects in the video, by
tracking each car in a selected region of interest. This is one of the
fastest methods to correctly identify, track and count a car object
with accuracy up to 78 percent.
In this research work, we explore the vehicle
detection technique that can be used for traffic surveillance
systems. This system works with the integration of CCTV cameras
for detecting the cars. Initial step will always be car object
detection. Haar Cascades are used for detection of car in the
footage. Viola Jones Algorithm is used in training these cascade
classifiers. We modify it to find unique objects in the video, by
tracking each car in a selected region of interest. This is one of the
fastest methods to correctly identify, track and count a car object
with accuracy up to 78 percent.
 iv
1. INTRODUCTION

Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management
I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.

Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.
Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu
Abstract— In this research work, we
explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.

Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.
Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu
Abstract— In this research work, we
explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.

Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.
Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu
Abstract— In this research work, we
explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.

I
Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management
I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.

Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.

m
Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu
Abstract— In this research work, we
explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.

Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.
a
Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.

Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management
I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.

g
Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.
Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu
Abstract— In this research work, we
explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.

e
Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management
I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.

processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.
Driver Assistant System is designed to assist drivers in the perception of any
dangerous situations before, to avoid accidents after sensing and understanding
the environment around itself. To date there have been numerous studies into
the recognition. Traffic accidents have become one of the most serious
problems. The reason is that most accidents happen due to negligence of the
driver. Rash and negligent driving could push other drivers and passengers in
danger on the roads. More and more accidents can be avoided if such dangerous
driving condition is detected early and warned to other drivers. Most of the
roads, cameras and speed sensors are used for monitoring and identifying
drivers who exceed the permissible speed limit on roads and motorways. This
simplistic approach, and there are no restrictions. II. BACK.
Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu
Abstract— In this research work, we
explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.

Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.
Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu
Abstract— In this research work, we
explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.

Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.
Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu
Abstract— In this research work, we
explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.

I
Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management
I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.

Image
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management

I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.

mage
processing
based Tracking
and Counting
Vehicles
Mr. Nikhil Chhadikar, Ms. Priyanka
Bhamare, Mr. Krushna
Patil, Mrs.Sangeeta Kumari
Vishwakarma Institute Of
Technology,Pune
nikhil.chhadikar15@vit.edu

Abstract— In this research work, we

explore the vehicle
detection technique that can be used
for traffic surveillance
systems. This system works with the
integration of CCTV cameras
for detecting the cars. Initial step will
always be car object
detection. Haar Cascades are used for
detection of car in the
footage. Viola Jones Algorithm is used
in training these cascade
classifiers. We modify it to find
unique objects in the video, by
tracking each car in a selected region of
interest. This is one of the
fastest methods to correctly identify,
track and count a car object
with accuracy up to 78 percent.
Index Terms—Digital Image
Processing, Automatic Traffic,
Computer Vision, Haar-like Feature,
Cascade Classifier, Vehicle
Detection, Traffic Management
I.

I NTR ODU CTION

In this research paper, we will
discuss about the unique car
identification and tracking in a
selected region of interest with
most accurate results. Vehicle
detection and counting play an
important part of many systems that
help to manage and control
traffic in cities. Main objective is
to detect and count the cars
with maximum accuracy and to
be able to do so on roads,
highways and in small lanes etc.
Our method uses foreground
objects, i.e haar cascades to detect
the cars, which takes input as
video or an image and processes it
to give the accurate count
of vehicles seen in it[10].
A video or live video footage is
divided into frames. These
frames were transformed into gray
frames and these gray frames
were given as an input to the system.
Then a specific region was
selected as a region of interest. By
using Haar-like features car
was detected. The car was tracked till
it goes out of the region of
interest. Every frame is compared
with the previous frame, if the
car is present in both the frames and
difference in their x and y
coordinates is less than max
(Width, Height) pixels then we
consider it as a same car. If the
difference is more than max
(Width, Height) pixels, then we
consider them as 2 separate
cars.
 3. THEORY

Nowadays, demands on vehicles are increasing because of the rapid population

growth. This leads to the major causes of higher road accidents. Nearly 3,500
people die on the road every day. It is found that the lane changing is the cause
of most accidents. The advantage of LDWS should be an optional tool to relieve
such disaster. LDWS is one of the methods that are proposed to solve the
problem. LDWS
uses the image processing mechanism to detect the lines of the image, selects
the current lane markers, warns the driver when the vehicle leaves the current
lane and tracks the lane marker for the next frame.
In this article, overviews of SHT and RHT are presented and the algorithms are
briefly introduced and
compared.

To solve the problems, a lane detection is proposed in this article as outlined by

the flow diagram.

Lane boundary detection is divided into 3 parts. The image that obtained from
the camera has many ingredients. The image preprocessing is applied to make
it appropriate for processing. Image preprocessing is performed on the road
image to locate the left-right lane edges by separating the sky
region and analyzing road region known as the region of the interest (ROI) in
this article to extract the
prominent road features such as lane markers. The information of the vanishing
point is used to limit the area in line detection and refer the point in RHT. Lane
marker detection part attempts to identify the lane marker from the detected
lines.

Lane Boundary Detection

Lane boundary detection consists
of 3 parts, i.e. image
preprocessing prepares a proper
image for further
processing from the original
image as shown in figure 2, line
detection to extract the lines
contained in
the images and lane marker
detection to detect the current
lane markers.
Lane boundary detection consists
of 3 parts, i.e. image
preprocessing prepares a proper
image for further
processing from the original
image as shown in figure 2, line
detection to extract the lines
contained in
the images and lane marker
detection to detect the current
lane markers.
Lane boundary detection consists
of 3 parts, i.e. image
preprocessing prepares a proper
image for further
processing from the original
image as shown in figure 2, line
detection to extract the lines
contained in
the images and lane marker
detection to detect the current
lane markers.
Lane boundary detection consists
of 3 parts, i.e. image
preprocessing prepares a proper
image for further
processing from the original
image as shown in figure 2, line
detection to extract the lines
contained in
the images and lane marker
detection to detect the current
lane markers.
Lane boundary detection consists
of 3 parts, i.e. image
preprocessing prepares a proper
image for further
processing from the original
image as shown in figure 2, line
detection to extract the lines
contained in
the images and lane marker
detection to detect the current
lane markers.
Lane boundary detection consists
of 3 parts, i.e. image
preprocessing prepares a proper
image for further
processing from the original
image as shown in figure 2, line
detection to extract the lines
contained in
the images and lane marker
detection to detect the current
lane markers.
Lane boundary detection consists of 3 parts, i.e. image preprocessing prepares a
proper image for further
processing from the original image as shown in figure 2, line detection to
extract the lines contained in
the images and lane marker detection to detect the current lane markers.
Lane boundary detection consists of 3 parts, i.e. image preprocessing prepares a
proper image for further
processing from the original image as shown in figure 2, line detection to
extract the lines contained in
the images and lane marker detection to detect the current lane markers.
Lane boundary detection consists of 3 parts, i.e. image preprocessing prepares a
proper image for further
processing from the original image as shown in figure 2, line detection to
extract the lines contained in
the images and lane marker detection to detect the current lane markers.
Lane Boundary Detection
Lane boundary detection consists of 3 parts, i.e. image preprocessing prepares a
proper image for further
processing from the original image as shown in figure 2, line detection to
extract the lines contained in
the images and lane marker detection to detect the current lane markers.

2.1. Image preprocessing

The proposed image preprocessing includes the grayscale image conversion,
image filtering, image morphology , and sky region removing which applies
some technique to remove some part of the image outside the ROI. This can
reduce the computational time in the vanishing point
detection method. In this article, the sky region removing removes the upper
half of the image. Consuming some part of the road surface is unnecessary
because this method focuses on the
lane markers.
Lane boundary detection consists
of 3 parts, i.e. image
preprocessing prepares a proper
image for further
processing from the original
image as shown in figure 2, line
detection to extract the lines
contained in
the images and lane marker
detection to detect the current
lane markers.
Lane boundary detection consists
of 3 parts, i.e. image
preprocessing prepares a proper
image for further
processing from the original
image as shown in figure 2, line
detection to extract the lines
contained in
the images and lane marker
detection to detect the current
lane markers
Lane boundary detection consists
of 3 parts, i.e. image
preprocessing prepares a proper
image for further
processing from the original
image as shown in figure 2, line
detection to extract the lines
contained in
the images and lane marker
detection to detect the current
lane markers.

2.2. Line detection

The line detection part consists of

the edge detection that includes a
variety of mathematical methods
that aim at identifying points in a
digital image at which the image
brightness changes sharply as
shown
in figure 5. Many edge
enhancement algorithms have
been proposed. The Sobel edge
detection is used
for this article because it is better
than Canny edge detection in
term of being less time-
consuming [5].
After edge detection method, the
image contains only white and
black pixels.

The line detection part consists of

the edge detection that includes a
variety of mathematical methods
that aim at identifying points in a
digital image at which the image
brightness changes sharply as
shown
in figure 5. Many edge
enhancement algorithms have
been proposed. The Sobel edge
detection is used
for this article because it is better
than Canny edge detection in
term of being less time-
consuming [5].
After edge detection method, the
image contains only white and
black pixels.

The line detection part consists of

the edge detection that includes a
variety of mathematical methods
that aim at identifying points in a
digital image at which the image
brightness changes sharply as
shown
in figure 5. Many edge
enhancement algorithms have
been proposed. The Sobel edge
detection is used
for this article because it is better
than Canny edge detection in
term of being less time-
consuming [5].
After edge detection method, the
image contains only white and
black pixels.

The line detection part consists of

the edge detection that includes a
variety of mathematical methods
that aim at identifying points in a
digital image at which the image
brightness changes sharply as
shown
in figure 5. Many edge
enhancement algorithms have
been proposed. The Sobel edge
detection is used
for this article because it is better
than Canny edge detection in
term of being less time-
consuming [5].
After edge detection method, the
image contains only white and
black pixels.
The line detection part consists of the edge detection that includes a variety of
mathematical methods that aim at identifying points in a digital image at which
the image brightness changes sharply as shown in figure 5. Many edge
enhancement algorithms have been proposed. The Sobel edge detection is used
for this article because it is better than Canny edge detection in term of being
less time-consuming . After edge detection method, the image contains only
white and black pixels.
The line detection part consists of
the edge detection that includes a
variety of mathematical methods
that aim at identifying points in a
digital image at which the image
brightness changes sharply as
shown
in figure 5. Many edge
enhancement algorithms have
been proposed. The Sobel edge
detection is used
for this article because it is better
than Canny edge detection in
term of being less time-
consuming [5].
After edge detection method, the
image contains only white and
black pixels.
The line detection part consists of
the edge detection that includes a
variety of mathematical methods
that aim at identifying points in a
digital image at which the image
brightness changes sharply as
shown
in figure 5. Many edge
enhancement algorithms have
been proposed. The Sobel edge
detection is used
for this article because it is better
than Canny edge detection in
term of being less time-
consuming [5].
After edge detection method, the
image contains only white and
black pixels.

The line detection part consists of

the edge detection that includes a
variety of mathematical methods
that aim at identifying points in a
digital image at which the image
brightness changes sharply as
shown
in figure 5. Many edge
enhancement algorithms have
been proposed. The Sobel edge
detection is used
for this article because it is better
than Canny edge detection in
term of being less time-
consuming [5].
After edge detection method, the
image contains only white and
black pixels.

The line detection part consists of

the edge detection that includes a
variety of mathematical methods
that aim at identifying points in a
digital image at which the image
brightness changes sharply as
shown
in figure 5. Many edge
enhancement algorithms have
been proposed. The Sobel edge
detection is used
for this article because it is better
than Canny edge detection in
term of being less time-
consuming [5].
After edge detection method, the
image contains only white and
black pixels.

The line detection part consists of

the edge detection that includes a
variety of mathematical methods
that aim at identifying points in a
digital image at which the image
brightness changes sharply as
shown
in figure 5. Many edge
enhancement algorithms have
been proposed. The Sobel edge
detection is used
for this article because it is better
than Canny edge detection in
term of being less time-
consuming [5].
After edge detection method, the
image contains only white and
black pixels.

The line detection part consists of

the edge detection that includes a
variety of mathematical methods
that aim at identifying points in a
digital image at which the image
brightness changes sharply as
shown
in figure 5. Many edge
enhancement algorithms have
been proposed. The Sobel edge
detection is used
for this article because it is better
than Canny edge detection in
term of being less time-
consuming [5].
After edge detection method, the
image contains only white and
black pixels.
Standard Hough Transform. The SHT computation consists of
calculating the parameter and
accumulating the data in the accumulator, finding the local maxima which
represent line segments of
the image and extracting the line segments from the maxima positions. The
disadvantages of SHT are
its computational complexity and huge storage memory consumption.
It assumes all of the white pixels of the image to be the points and
converts them into ρ-θ plane. ρ is
the line connecting the polar coordinate to the origin where the x-axis intersects
the y-axis. θ is the angle
between ρ and x-axis as shown in figure 7. Each point in the x-y plane will have
the infinity lines that
pass the point. Those lines are converted by equation (1).
0 0
cos sinx y
 

, (1)
where 𝑥 and 𝑦 are the coordinates corresponding to the angles θ (𝑥 , 𝑦 ).
0 0 0 0
The line plot of those lines
will be the sets of point in the ρ-θ plane. So, the point connection will be the
sinusoidal wave for each
point in the x-y plane as shown in figure 8. The intersection of lines in ρ-θ plane
means they have the
same value of ρ-θ. By voting θ in the accumulator will get the set of points in
the x-y plane that has the
high possibility to be on the same line.

4. SOURCE CODE
5. Final Output(Result)
This research designs a classification
system to determine object
as specific type of vehicle. Haar
Cascade Classifier is used to
determine object as car and
counted the number of passing
vehicles on the specific road using
traffic videos as input. The
detection rate of this system is
affected by the scale factor
value, different scale factor value
providing varied detection
rates. In obtaining high detection
rate, the scale factor value
giving the best performance to
classifier should be determined. .
In the future, providing skillful
and robust vehicle detection
system will be a challenging task in
this field.
 6. Conclusion

This research designs a classification

system to determine object
as specific type of vehicle. Haar
Cascade Classifier is used to
determine object as car and
counted the number of passing
vehicles on the specific road using
traffic videos as input. The
detection rate of this system is
affected by the scale factor
value, different scale factor value
providing varied detection
rates. In obtaining high detection
rate, the scale factor value
giving the best performance to
classifier should be determined. .
In the future, providing skillful
and robust vehicle detection
system will be a challenging task in
this field.
Lane detection has been presented in this article image preprocessing, line
detection and lane marker
detection. It has been shown that the main advantage of the RHT method over
the SHT is computational
time. The difference in computational time between SHT and RHT with 5
threshold is about 5.1 percent,
RHT with 10 threshold is about 4.99 percent, RHT with 15 threshold is about
2.73 percent and RHT
with 20 threshold is about 0.06 percent, according to table 1.
The system still has two further parts to complete, i.e. the warning part that
finds the condition of
lane changing and the tracking part that reduces the computational time of the
current frame using the
information of the previous frame.

In the future work, the developer

can develop the system into the
application on a smartphone,
which
can be attached to the car front
REFERENCES

Pl: Note
Any sentences, paragraphs, photos, sketches, models, flow sheets etc that are taken from any
text book, journal, web page should be properly referenced both within the body of the
Report (wherever it occurs) as well as at the end in The references section. In sequence use
1, 2, 3, 4, …n for the reference within the body as within brackets such as (1), (2), (3), (4),...
(n) immediately after that sentence, paragraph, photos, sketches, models, flow sheets etc. In
the reference section, which is this page, in sequence of appearance within the body (or text)
of the Report list those References. These reference numbers should match with the actual
reference used and reference cited for any material used. Format for the References would be.
(Remove this Paragraph from this Page when you prepare References for this Section that
you had used to prepare the Report)

(1) Raman Maini and Dr. Himanshu Aggarwal 2009 IJIP 3 pp 1 – 12.

(2) Pekka Kultanen, Lei Xu and Erkki Oja 1990 IEEE 1 pp 631 – 635