International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056

Volume: 04 Issue: 05 | May -2017 www.irjet.net p-ISSN: 2395-0072

Traffic Signal Duration Control by Estimating Vehicle Density

Nithika Bhasi 1, Chithra V2
12Assistant Professor, Department of ECE Dhanalakshmi Srinivasan College Of Engineering Coimbatore, India
---------------------------------------------------------------------***--------------------------------------------------------------------
Abstract In India traffic congestion is becoming a very It is annoying and effects quality of life for everyone who are
serious issue day after day. The heavy traffic jam is a result of stuck in it
large amount of vehicle which is caused by huge population It causes many problem and limits future economic
and development in economy. As a part of globalization, time expansion.
consumption and management means a lot. The fast moving
Cars stuck in congestion produce more CO2 per mile
world in which we live today does not want to wait at any cost
than the
in middle of the time consuming traffic signals. As a result of
this mentality of the people, safety measures are discarded free-flowing traffic.
and violated intentionally. Thus a need for density based time It adversely affects the people who are in emergency.
saving traffic signal control unit aroused. The density based When compared with the growth of population,
system would be better than the time based system control as vehicle growth is also increasing rapidly which is influencing
it scans the density of vehicles on road based on Digital image traffic system. The normal function of traffic lights should be, to
processing. More accurate and easily maintainable system is be more controlled and coordinated to ensure that traffic
what we need. This time saving control system can be used in moves as smoothly and safely
all frequently congested junctions. The proposed system has
as possible. Different control systems are used for this, such
two sections; the traffic section and the control section. The
vehicle density in the junction is estimated by capturing the as simple clockwork mechanisms, to computerized control and
traffic condition using digital cameras and processed by coordination systems that are used to minimize delay for
MATLAB programming. Based on the captured and processed people using the road.
information on Vehicle density the signal durations can be One of the methods used for controlling the traffic
increased or reduced using microcontrollers. signal was the use of timers [1]. It presets the time in the traffic
signal for the red light and green light. It works according to
Index Terms- traffic density estimation, traffic control, the timing and not depending on the traffic density. The main
control section, traffic section, PIC16F87XA, ostus algorithm.. drawback of this method is it does not monitor the traffic
density and it does not control the traffic signal timing which
1. INTRODUCTION causes the vehicles to wait on the junction for hours till the
signal becomes green. Manual controlling of traffic signal is one
The traffic signal system in almost all cities in India is of the oldest technique which requires man power to control
based on the time, that results in wastage of time, more fuel the traffic [5]. At different countries, traffic polices are allotted
consumption and more congestion on roads. When the vehicle for a particular area or city to control the traffic manually.
on the road increases the speed of traffic stream decreases These traffic polices carry different devices to control the
which results in some congestions. This makes the whole traffic such as sign board, sign light and whistles. The main
population in the cities to stand still in the middle of the heavy drawback of this method is it becomes difficult for a traffic
traffic jam daily. Some steps or measures have to be taken to police to control large amount of traffic in a four way junction
overcome this situation. This paper is proposing a method to as a result of which congestion increases.
estimate vehicle density using digital image processing and
controlling the traffic signal. Traffic congestion near the traffic Another technique used for traffic controlling was with
signal is one of the main problem which we are facing in our the help of sensors [6]. The controller uses the input from
day to day life, mainly in metropolitan cities. Traffic congestion detectors, that are sensors that inform the controller that,
whether a vehicle or road users are present, to adjust signal
are causing challenges for all developed and developing cities.
This leads to so many problems such as waiting in front of the timing according to the limits set by the controller
signal for hours for the green signal which leads to time programming. It can provide more time to a junction that is
wastage. It even creates problem for emergency vehicles like experiencing heavy traffic,. These detectors are of three types
ambulance, fire force etc. Cities are mainly affected by it which they are, in-pavement detectors, non-intrusive detectors, and
motorized road users detector [6]. These detectors are buried
results in massive delays, increased fuel wastage and monetary
losses.Traffic congestion is a concern largely for four reasons: inside the roadways. Inductive detector loops are the detectors
which are commonly used. They are sensors which are buried
inside the road to detect the presence of traffic waiting at the

2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 321
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 04 Issue: 05 | May -2017 www.irjet.net p-ISSN: 2395-0072

signal, and thus it can reduce the time when a green signal is 2. EXSISTING SYSTEM
lighted on an empty road. During low traffic density a timer is
With the advancement in wireless communication, so
mostly used as default, and it will have a backup in case of
many techniques are developed in which vehicles can form a
sensor failure. Consequently, small vehicles and bicycles
network for exchanging traffic information among themselves.
with low metal content may fail to be detected which causes
For this a infrastructure less self maintaining traffic
them to wait for long time unless there is a default timer as
information system is used. In this system as the vehicle travels
part of the control system. This is the major drawback of this
they collect the traffic data by acting as mobile sensors. For
method.
traffic sensing smart phones are used as they are equipped
Another method is coordinated control which has 2 with a variety of sensors such as gyroscope, Accelerometer,
systems and they are synchronized signal system and GPS etc. These sensors can be used effectively for collecting
coordinated signal system [4]. In modern coordinated systems traffic data. GPS receivers are used mainly for many years.
the drivers can travel a long distance without encountering a However GPS receiver consumes a lot of power, and hence they
red light. This coordination can be done easily on one-way shortens the battery life [7]. So as a result of this the existing
streets with constant levels of traffic. One of the drawback of system uses accelerometers for sensing the traffic
this method is it monitors the speed of the vehicle according to information[1].
which the green light is encountered once the speed exceeds a
Traffic information system is very essential for solving
particular level the driver will arrive on a red indication and
a traffic problem effectively. Vehicle speed, vehicle density and
end up stopping, drivers traveling too slow will not arrive at
travel time which are the Real time traffic information, can help
the next signal in time and they cannot utilize the green
the motorists choose a route more wisely and easily if they
indication. Some other methods are fixed time control, adaptive
know it before. Instead of unnecessarily getting stuck on a
control etc. These are some of the oldest and the existing
congested road, many drivers can choose free route if the
methods used for traffic signal controlling.
information about the traffic condition is known before hand.
The proposed method is designed in such a way, that it Almost all the traffic information systems in India are based on
solves all these drawbacks and problems which are faced in some fixed infrastructure, where datas are collected from fixed
existing method. And they are solved mostly by estimating the sensors such as surveillance cameras and Inductive loop
density of vehicles waiting at the junction and accordingly detectors. it takes a great effort to install sensors in such kind
controlling the traffic signal. It can give more time to a junction of systems. In many cases, the road surfaces are digged to
that is having heavy traffic, or shorten the time at the junction implant the sensors underneath it, which requires a large man
that has few or no traffic waiting for a green light. In urban and power and lots of time. Not only that it is even much expensive
crowded cities its a regular scenario to see huge traffic to install sensors in a large area such as metropolitan cities.
congestion at the traffic signal which leads to so many Moreover, maintenance of these fixed sensors is also difficult
problems. The traffic density is estimated using digital image [1]. This system could be made less expensive and effective if
processing and using zigbee wireless transmission system the the vehicles themselves collect and distribute the traffic
data is transmitted to the control section, from where the information. The main drawback in this type of system is, how
duration of the signal is produced and passed to the traffic to collect and process the data as now the sensors are mobile
section using zigbee transmitter. and distributed. Among these issues, the main things every
system designers need to consider are: (i) which are the
The main benefits of the proposed method are:
sensors that must be used. (ii) how to use them properly, to
It Increases the traffic handling capacity of roads. effectively collect the traffic data, and the last and the most
Reduces collision, Reduces the waiting time for both important one is (iii) how to convert the data into information
vehicles and pedestrians. which can accurately convey the real traffic condition. The
It encourages traveling within the speed limit to meet information is gathered by undergoing 3 major steps they are:
green lights. Data collection, State classification and Regression based speed
It reduces fuel consumption by unnecessary stopping estimation.
and starting the vehicles, reduces noise and vehicular
A. Data Collection
wear and tear.
It reduces drivers frustration.
The main advantage is in reducing the duration of To collect the real data sensed by an Accelerometer
green light in the junction were the traffic density is low. In actual field experiments are conducted. For data collection
cases, where the junction will be empty and still the green light purpose a simple iOS application is developed and an iPhone5
will be lighted causes vehicles of other junction to wait without is used as a mobile device. The following application is
any reason, these problems can be solved using this. designed only for data collection. The main task of this
application is to record the status of the essential data and the
accelerometer which will later be used for analysis purpose.

2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 322
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 04 Issue: 05 | May -2017 www.irjet.net p-ISSN: 2395-0072

The experiment starts by, placing the mobile phone above the method as it collects all the information in database will
dashboard of a vehicle. The mobile phone can be placed any experience so many problems.
where and in any orientation as the initial state of the
accelerometer will be calibrated automatically by the
application.
B. State Classification

The data is divided in to blocks once they are collected,

each having T seconds. By using this the average speed of the
vehicle can be estimated over the period of T. Speed estimation
will be mainly based on the various activities of the vehicle.
The value chosen for T is 360 seconds. It is important to
identify when the vehicle is stationary at a period of T. It is
obvious that a vehicle can be always in any of the 2 states:
Fig 1. Average speed as a function of total mobile time.
stationary or mobile. These states, that is whether the vehicle is
mobile or stationary can be identified by the data sensed by 3.PROPOSED SYSTEM
accelerometer. To distinguish between the mobile state and
stationary state of a vehicle the following 2 points should be The fast moving world in which we live is facing
considered: 1) A vehicle will be considered to be in stationary severe problems due to increasing traffic congestion day after
state if the initial value which is obtained from the day. The main reason behind this increase in congestion is the
accelerometer is less than the threshold value. 2) At the same population expansion and its need for high volume of
time the vehicle will be considered to be in mobile if the vehicles. The traffic jams causes frustration to the drivers and
deviation from the initial value is greater than or equal to the results in road rage, because of which safety measures are
threshold value for a duration of at least T seconds discarded and violated intentionally. Thus a density based
time saving traffic signal control system is designed. The
C. Speed Estimation proposed technique of density based system would be better
than a time based system as it scans the density of the
Once the states are classified, that is vehicle is
vehicles on a road based on digital image processing. With the
stationary or not in a period T, the next task is to transform the
help of the new system more time can be given to light the
classified data into average speed. We have to correlate the
green signal at junctions which experience heavy traffic, and
total time which the vehicle was in motion during the period T
can even shorten the time at the junction which have few
to its average speed. If the vehicle can be in motion for a longer
traffic density. A more accurate and easily maintainable
period of time, then there are chances that the vehicle can
system is our main concern . Our system can be used in all
accelerate to a higher speed.
frequently congested road junctions. It has 2 main sections :
Next, the total time in which the vehicle was mobile is Control section and the traffic section.
mapped to the average speed. As this is the training phase, we
use the speed obtained from the GPS receiver for reference.
Once the mapping results are obtained, the average speed can
be estimated by using accelerometer data. On the basis of the
collected data, the average speed is calculated at every second.
The average speed of the vehicle which is denoted by vT, can be
obtained from the sampled datas. The total mobile time is
taken as 360 seconds, the average speed ranges from 30 to
85km/h. These experiments are performed mainly for sensing
the traffic condition in the particular area by the vehicles
themselves. With the help of which they can choose wisely a
more easy and less congested route. It is able to understand
that, If a vehicle is able to be in motion for a longer period of
time, then it means that the traffic condition is smooth or the
route is less or not congested. This enables the drivers to drive
Fig 2. Block diagram of proposed system
with a great speed. If for the time m= 360 seconds, the
vehicles move without any congestion and without any stop. There are mainly five units in this system they are:
This will effect the accuracy of the estimation process and
Microcontroller Unit.
makes the average speed estimation more difficult. This
Power Supply Unit.

2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 323
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 04 Issue: 05 | May -2017 www.irjet.net p-ISSN: 2395-0072

Display unit. The main blocks involved in the system are:

Communication Unit. Block 1: Camera
Software Unit.
Microcontroller is mainly used for controlling the Block 2: PIC 16F877A Microcontrollers
traffic signal and duration estimation .The condition about the
traffic density which is captured by using camera is processed. Block 3: Max232
It is processed through digital image processing and is send to
Block 4: Zigbee
the microcontroller at the control section. In the control section
the controller estimates the duration for which each signal is to
Block 5: LCD
be lighted on the basis of the received data about the density .
The microcontroller at the traffic section is used to control the The camera is used for capturing the images of the
traffic signal light. Here PIC 16F877A microcontroller is used. vehicles which are waiting at the traffic junction. Afterwards
Second, the Power supply unit working can be these captured images are processed by using digital image
explained. The supply of 230V AC is converted to 12V DC which processing.
is given to the system. For this purpose a step down The Zigbee module which is used acts as both
transformer is used. Supply is converted from AC to DC transmitter and receiver[3]. The transmitting and receiving
because the microcontrollers will support only DC supply, pins of 8051 controllers are connected to the same pins of
which is done by using bridge rectifiers. The rectifier output Zigbee respectively. The signal from the microcontroller is
will have so many ripples, so for filtering out those ripples transmitted serially to other controller at the traffic section by
2200uf capacitors are used. The filter output is given to the using Zigbee module via UART port. The Zigbee is a transceiver
7805 voltage regulator which will convert the 12V DC again that transmits and receives data or information
into 5V DC. The output of the regulator will be filtered by using simultaneously. It can transmit data to a wide range. These
the capacitor. As a result of this we will get a pure 5V DC as blocks altogether computes the proposed system.
power supply unit output. Here microcontroller which we are
using is capable of using the supply of 5V DC. Power supply The main operation or the working of our proposed
unit is very important for the proper and effective working of method is:
our system. The camera is placed on the top of the traffic signal to
The display unit used here is the LCD (Liquid crystal obtain the clear view of vehicles present at the junction. The
display). A liquid crystal display (LCD) is a flat panel display, traffic level is estimated before entering the traffic section by
which have flat screens used for displaying graphics and using MATLAB. The traffic signals are controlled by two
contents virtually. It uses the properties of liquid crystals (LCs), methods, automatic and manual method. In automatic mode
that is the light modulating property. In this proposed method the camera is used to capture the traffic image and the data is
LCD is used to display sensor value, which contain the details processed by using MATLAB process to identify the vehicle
about the traffic light signal and the durations. density. These information, that is traffic level is transmitted to
the control section through zigbee. The estimation of timing
Zigbee is the communication unit which is used to duration and controlling is performed by the microcontrollers.
transmit the sensor values to the monitoring section. Zigbee is Based on the traffic density, green signal can be lighted for
a low cost, low power and low latency wireless communication more time on the road which has the more number of vehicles
standard. It is IEEE 802.15.4 standard for low cost WPANS. As and the congestion on the roads can be reduced. At the same
it is having low latency it reduces the average current. Zigbee time green signal can be lighted for short period of time, if the
operates in the ISM radio bands that is (Industrial, scientific density of the vehicles at the junction is less. Thus the traffic
and medical) which is 2.4GHz.Its data rate varies from 20kbit/s signal lights are automatically controlled based on the density
to 250kbit/s. of vehicles on corresponding roads.
The software unit in the method used are: Proteus and 4.EVALUATION AND RESULT
MATLAB. Proteus is a stimulation software where the
microcontroller stimulation is performed. The program which The proposed scheme is simulated on MATLAB and
is debugged in the controllers are executed in the MpLab. In the Proteus. The capturing and processing of traffic image is
MATLAB the digital image processing operation takes place to performed in digital image processing domain, and the
estimate whether the traffic is heavy or low. Otsus algorithm is stimulation of image processing is done in MATLAB. Firstly, the
used here. image path and filename is selected and a particular image is
fetched to determine whether the image density is heavy or
low. The image is colored which have the primary colors Red,
Green and Blue. The algorithm used for this stimulation is

2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 324
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 04 Issue: 05 | May -2017 www.irjet.net p-ISSN: 2395-0072

Otsus algorithm. The RGB color is an additive color in which stimulation is the duration of time for which the green and red
red, blue, and green light are added together in various ways light must be lighted, and controlling of the traffic signal.
to produce a broad variety of colors. The main purpose of
RGB color model is sensing, representation, and displaying of
images in electronic systems, such as computers and
television.
The RGB color model is converted to hue which is hue
saturation value. Hue is the determination of pure color.
Saturation is degree at which the pure color is diluted with the
white light. Hue and saturation is together called color carrying
information in an image. It separates the intensity component
from the color carrying information. In order to determine the
pixel value of the image accurately first the image is converted Fig 5. Simulation of traffic section
to grayscale image and then to binary image. The graphical
representation of image to determine each color intensity is 5.CONCLUSION AND FUTURE SCOPE
histogram. In this paper, instead of using the old methods for
controlling the traffic signal, a new method is proposed in
which the traffic is controlled by estimating the density of
vehicles present in the junction. This proposed method helps
in reducing the problems related to the heavy traffic
congestion at the traffic signal junction which is continously
increasing with the passing days. The present world is too
busy that it doesnot want to waste there valuable time in the
middle of crowded junctions. As a result of this the proposed
system is introduced which is a time saving traffic signal
control unit, based on density. This technique will be better
than the time based system. This paper covers the
methodology of implementing the traffic signal control
system based on the density of vehicles on road. The main
Fig 3. Binary image of traffic density objective of this paper is to estimate the density of vehicles
using digital image processing and embedded system. The
paper thus gives a method to reduce congestion on road,
saves fuel and valuable time.
In the future work the information about the traffic
density and the present traffic signal at the junction can be
send to the vehicles with the help of GPS and Zigbee
trannsreceiver. So that instead of getting stuck unnecessarily
on a congested road, drivers can be able to take an alternative
less congested route if the information about the road traffic
condition is available to them in advance.
REFERENCES
Fig 4. Simulation of control section [1] Amrita Rai and Govind Singh Patel. Muliple Traffic
The gray threshold computes a global threshold level Control Using Wireless Sensor and Density Measuring
that can be used to convert an intensity level image in to a Camera Sensors & Transducers Journal Vol.94, Issue
binary image. The image is converted to binary mainly in order 7, July 2008, pp. 126-132.
to find a region of Interest. It is easier to determine the pixel [2] H.R Kashani and G.N.Saridis Intelligent Control for
value of an image from binary image then from a colored Urban Traffic Systems International Federation of
image .The processed information is transferred to proteus Automatic Control Automatica, Vol.19, May 2014.
simulation through serial port. If the output is 1 it is considered [3] L. Wu, Experimental study on vehicle speed
to be heavy traffic and if it is 0 then the traffic condition is estimation using accelerometer and wheel speed
considered low. The proteus program is processed in Mplab measurements, in Proc. IEEE International
and debugged in the microcontroller. The output of this Conference on Mechanic Automation and Control

2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 325
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 04 Issue: 05 | May -2017 www.irjet.net p-ISSN: 2395-0072

Engineering, Jul. 2011.

[4] Manoj Kanta Mainali & Shingo Mabu Evolutionary
Approach for the Traffic Volume Estimation of Road
Sections IEEE pp100-105, May 2010.
[5] Mohit Dev Srivastava, Prerna, Sumedha Sharma,
Smart Traffic Control System Using PLC and SCADA
International Journal of Innovative Research in
Science, Engineering and Technology Vol.1, Issue 2,
December 2012.
[6] Mrs. Ligi K, Vehicle Density Estimation using GPS and
Zigbee, International Journal of Engineering
Research & Technology (IJERT), Vol. 3 Issue 3, March
2014.
[7] Ms Promila Sinhmar, Intelligent Traffic Light and
Density Control Using IR Sensors International
Journal of Advanced Technology & Engineering
Research(IJATER), Vol.2, Issue 2, March 2012.
[8] R. Bhoraskar, N. Vankadhara, B. Raman, and
P.Kulkarni, Wolverine: Traffic and road condition
estimation using smartphone sensors, in Proc. IEEE
International Conference on Communication
Systems and Networks, Jan. 2012.
[9] Sooksan Panichpapiboon, Member, IEEE, and
Puttipong Leakkaw,Traffic Sensing Through
Accelerometers,June 2015.
[10] Vivek,Tyagi, Senior Member IEEE, Shivakumar
Kalyanaraman, Fellow, IEEE, and Raghuram
Krishnapuram, Fellow, IEEE Vehicular Traffic
Density State Estimation Based On Cumulative Road
Acoustics in IEEE Transaction On Intelligent
Transportation System. Vol.23. No.23 September
2012.
[11] Wanjing MA and Xiaoguang YANG Design and
Estimation of an Adaptive Bus Signal Priority
System Based on Wireless Sensor Network
Proceeding of the 11th International IEEE
Conference on Intelligent Transportation Systems
University Science, 1989.

2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 326