Review Paper on Intelligent Traffic Control system using Computer Vision for Smart City

International Journal of Scientific & Engineering Research Volume 8, Issue 6, June-2017 14
ISSN 2229-5518
IJSER © 2017
http://www.ijser.org
Review Paper on Intelligent Traffic Control system
using Computer Vision for Smart City
Janak Trivedi1
, Dr. Mandalapu Sarada Devi2
, Dave Dhara3
1
Asst. Prof. - Electronics & Communication Department- G.E.C. Bhavnagar, G.T.U., 2
Principal. – A.I.T., Ahmedabad, G.T.U.,
3
Asst. Prof. - Electronics & Communication Department- G.E.C. Bhavnagar Gujarat, India
1
trivedi_janak2611@yahoo.com, 3
saradadevim1@gmail.com, 3
dave.dhara24888@gmail.com
Abstract— In today scenario city will try to modify in the form of smart city with better facilities in terms of education, social-economic life,
better transportation availability, noise free – Eco-friendly environment availability, and ICT- Information and communication technology
enabler for development in the city. In this paper, we are reviewing different work already done or draft by some research in the field of traffic
control system – for better monitoring, tracking and managing using a computer vision system. Nowadays, most of the city installed with
C.C.T.V. – camera for monitoring the traffic related activity.
Keywords— Transportation Management System (TMS), CCTV camera, Gaussian mixture model, Video, Image processing, smart city, parking
system, smart phone, Vehicle Tracking
1. INTRODUCTION
The Currently Indian government is trying to develop smart
cities and already announces 3 stages, in which nearly 60 cities are
selected. In the near future the list of the smart cities will further
increase [A]. First list displayed by government mentioning 2 cities
from Gujarat – Ahmedabad and Surat. Urbanization increases
rapidly within the last few years. In this intelligent transportation
system or traffic management in real time must be done with
available resource to avail better and better lifestyle for everyone
with less congestion of traffic so less noisy environment, less time
required to reach one place to another place, proper parking
availability etc. Currently, many of researchers trying to work on
developing algorithms for solving traffic related issues.
This paper further goes on with discussing - what is smart city?,
Challenges under smart city mainly due to traffic control systems in
Section II for motivation of research, then moves to discussing
section III with different algorithms discussed in various research
papers, Section IV discussed about smart phone or android
application useful for traffic monitoring, Section V gives a
conclusion and finally give some idea about future works or what
will be done for better implementation of traffic management in
smart city?
2. SMART CITY
A city can be defined a smart when investment in human and
social capital, traditional (transport) and modern (ICT),
communication, infrastructure, fuel, sustainable economic
development and a high quality of life with a wise management of
natural resource through participatory action and engagement. Smart
city is an urban space that tries to improve the daily life of all
citizens (with the knowledge of their habits, experiences, culture,
education, behaviour and different facilities available. Smart City is
capable of sensing what is happening in the city – real time parking
availability, traffic management, route navigation, intelligent
transportation system, availability of beds in hospitals, energy
consumption, water and air quality, temperature, noise, solid waste
management, security, transparency and accountability etc.
A. Smart city infrastructure
Smart city infrastructure divided along 4 parts [B] as shown in
the fig. 1 and from that our discussion more focusing on Physical
infrastructure- in that transportation module.
Fig. 1 Smart city infrastructure [B], from that our research of
interest is transportation module.
B. Smart city example
1. Iowa: smart water meter, can compare with neighbours
usage.
2. Dublin (Ireland): IBM’s “ParkYa” app for finding a
parking slot.
3. London: Mayor gets constant data analysis of twitter feeds.
4. Surat: online monitoring of water quality and much more.
C. Traffic related problem
Here some problem related to traffic is mentioned, which will be
solved for real time traffic monitoring, and these are…
• Creating online tool for traffic management.
• Selection of hardware platform and its sensor drivers,
proper collection and synchronization of different sensors.
IJSER

ISSN 2229-5518
IJSER © 2017
• Smart applications like mobile applications for smart cities,
smart mobile application for connecting driver and
passenger, humidity, pollution, traffic congestion
monitoring.
• Increasing safety, comfort, economy of the citizen, while
also providing new entertainment options.
• Provide capabilities for real-time sensing, signal and
multimedia processing, and sharing of information with
other users and server applications, like real time
transportation information sharing with all other
stakeholders.
• Real time vehicle tracking system under challenging light
conditions.
• Real time parking availability for vehicles in smart cities.
• Intelligent traffic light system with the use of different
sensors, especially in those junctions where maximum
traffic increased.
• Reduce traffic congestion to avoid pollution, accident like
situation.
• Management requires in transportation, that every
employee, whose duties are at fixed times, should avoid
their personal vehicle, but use public transportation to
reduce the amount of pollution and traffic congestion
places.
• Novel approaches required to implement intelligent
transportation systems within smart cities using a computer
vision system.
3. DIFFERENT METHODS ALREADY IMPLEMENTED
The approaches described by, Benjamin Coifman and his team
members [1] for vehicle tracking using four steps. Firstly automated
segment vehicle from background, detection of all types of road
vehicle and vehicle detection under challenging light conditions,
that’s second and third stage and lastly operate in real time. But
according to their description all vehicle tracking in different light
condition is not up to the mark. They also discussed different
algorithms for vehicle tracking, like model based tracking, region
based tracking with the use of Kalaman filters, active contour based
tracking and feature based tracking with the use of Kalaman filtering,
normalized correlation and Texas instruments C40 digital signal
processor chips.
Mohamed S. Shehata [2] and his team members describe
difficulties for detection of shadow, snow, rain and glare. And from
that snow detection is the most difficult task. In such incident
detection most common problem is wrongly detection not actually
presents that particular incident. They have discussed method based
on pixel’s brightness, normalized red, green and blue concept, hue-
saturation-value (HSV) color space concept. And then concentrated
on Gaussian mixture model with EM method, then use of Kalaman
filtering for better detection of certain incidents. There is difficulty in
finding rain road incidents. In that case morphological operations
required. Similar manner glare detection can also be focused as a
future research area.
Authors of [3] explain their arguments on Parking, Deposit
System (PDS) as an alternative arrangement for road pricing as a
study case. They had collected some information based on, basic
questionnaire related to traffic and parking with end users, based on
that designed study case with user benefit in terms of implementation
of PDS. Smart Parking further discussed in [4] based on wireless
networks and sensor communication. In that hardware architecture
implementation with the help of wireless transceivers, parking belts,
computer, infrared devices etc. And software architecture developed
with driver module, communication module, function module and
application module. This implementation is done with different
reaction scheme steps. Calculation can be done with the help of
binomial distribution, Poisson distribution with its mean. Simulation
result verified using Matlab version.
From, Faculty of engineering and architecture, Kore University of
Emma [10], gives architecture for vehicular parking in smart cities –
Intelligent Parking assistants (IPA) with 4 modules. (1) User
interface module (2) Communication module (3) Functional module
(4) Parking space control module. IPA in preliminary stage and it
can be modified further by including a smart phone.
Static and dynamic traffic assignment (DTA) problem discussed
in [5] with concluding that average travel time under a pre-time
signal policy is better than time actuated signal policy. Normally for
traffic related computation research based on a static assumption not
considers temporal flow distribution, so equilibrium with signal
setting and traffic assignments are required [5]. For that bi-level
framework theory required to solve problem related DTA. In bi-level
framework upper level and lower level solve different problems
related to DTA. For achieving Equilibrium DTA, Mathematical
based model, iterative Heuristic algorithm, Genetic algorithms,
Webster’s formulation; game-theoretical methodology, Mean
absolute percentage error (MAPE) and root mean square percentage
error (RMSPE) are explained.
Other papers [6] [7] gives attention towards public transportation
service. That is new thinking towards decreasing traffic congestion
as well as noise and sound pollution generated by personal vehicle.
Anutosh maitra [7] and his team members gives new idea for
controlling traffic, in the sense that more and more awareness
regarding public transportation uses for daily office users, or daily
passengers in real time. To achieve this, a description based on (1)
Commuter views (2) Vehicle operator/Driver View and (3) Control
centre view all these 3 works together. To implement this scheme
proper management with all 3 views required at time to time and
perfectly. And in [6] location of all the bus stops with its current
position and route accurately if available, to end user, may be most
of the people give their first choice to select public transportation.
For that 2 different algorithm discussed (1) Initial mapping algorithm
(2) Map Algorithm.
Although from earlier discussion public transportation more
healthy in terms of environment and traffic management, currently
more and more vehicle – two and four wheeler are shown in daily
life. For that scenario smart phone based traffic control management
gives better option. This thing discussed in [8] with SMaRTCaR.
Discussion based on, General packet radio service (GPRS),
Universal mobile telecommunication system (UMTS), and Vehicular
ad-hoc Network (VANET), an Arduino board.
The minimum time required to reach the destination from the
source with particular routes gives attention of many researchers
from industry as well as academic. As per [9] different vehicle
routing algorithms are Dijkstra’s algorithm, A* algorithm, Tabu
search, Ant based colony Genetic algorithm, Hybrid genetic
algorithm. In this paper comparison of these algorithms, with a scope
of improvement discussed with use of Simulation of Urban Mobility
(SUMO) and Traffic control interface (TRACI). Real time
implementing gives many benefits, Likes Company who wants to
take and give orders on time to the customers, serious life can be
saved by saving time, etc.
Two different types of congestion [11] are (1) recurrent and (2)
Non-recurrent congestion. Some open challenges related to traffic
IJSER

ISSN 2229-5518
IJSER © 2017
management system are- Data Fusion, processing and aggregation
(DFPA), lack of traffic parameter accuracy discussed. To solve that
problem wireless sensor, machine to machine (M2M)
communication, use of cellular/3G/LTE Networks, multimedia
oriented social area network (MMSN), fuzzy rough concept, novel
fusion technique based on yegar’s formula, neural network concept
discussed.
Said alabdallaoui and [12] his teammate indicate, Random early
detection (RED) approach in some heavy traffic congested area or
junction, using the concept of (1) Average queue size and (2) Pocket
dropping probability. RED explained by four phases (a) Learning
phase and traffic estimation (b) setup phase (c) Calculate real load
for each lane (d) RED algorithm application. Here major problem
implementation complexity and its deployment costs.
Another initial stage research going on intelligent transportation
by Fenghua [13] and his team members from IEEE explained
parallel transportation management in smart cities based on artificial
systems, computational experiment and parallel execution (ACP)
approach.
Gitae, Yew soon [14] and their team members try to find a
solution of dynamic vehicle routing problems under traffic
congestion, with the help of Markov decision process Model.
Dijkstra’s algorithm and Bellmen’s iteration scheme for finding
shortest path also explained by their limitations. And finally roll-out
algorithms with approximate dynamic programming (ADP) and
Neuro-dynamic programming method for better solution explained
with a case study.
4. SMART APPLICATIONS
Nowadays all people used smart phone for not only
communication purpose but also for entertainment, for social media
interaction, education, health tips and many more. If we combine a
smart phone with vehicle together make smart transportation system.
Smart phone support GPS location, Wi-Fi availability, in building
the blue tooth device and many more things are very much useful for
better transportation management.
For maximum use of public transportation, GBUS [6] consist
mobile application for data related to bus stop and its route
respectively, build a route from the collection and synchronization
using map- matching and string matching algorithms.
My Ford mobile application [8] useful for monitoring of electrical
vehicles, charging status of electrical vehicles, other electric vehicle
charging station, etc. The other Smart-phone application discussed in
same paper which can be able to detect and notify accidents, for that
smart phone camera also useful for awareness about accident
situation. TOMTOM and GRAMIN [11] systems are useful for
drivers, due to their services regarding location, shortest route
towards destination, etc. Easy tracker [6] systems are developed for
tracking, mapping and measure different parameter in the context of
traffic management. For location, purpose we can use Google maps
and Map Quest such like applications.
FastPrk [11] application uses M2M technology, for servicing fast
parking in certain areas. Toyota developed- Toyota friend [8], private
social network for Toyota car owners for their facilities. And so
many other companies also trying to make better and better
application for transportation management system (TMS) like –
sharing your personal details with your vehicle, with other groups
and members so maybe, if they want to wish to join they can join
you. So in this way also we can reduce use of traffic and that is also
one great step moment towards TMS services.
5. DIFFERENT PROJECT AND COMPANIES WORKED IN TMS SERVICE
Nowadays Indian government keep quite interested, in
developing smart city in INDIA. So naturally industry hopes their
some eyes on gaining something from it, with developing some
interest of area to get fruit from smart city structure. Better
transportation management, system availability is one of the highest
interests of the area. Here from different research papers listed some
companies or project, which are already running their system for
better TMS service.
With the help of video-image processing system – TRIPWIRE [1]
follow the movement of vehicles in terms of time respect to another
vehicle. Commercial TRIPWIRE systems are: AUTOSCOPE,
CCATS, TAS, IMPACTS. Other systems developed with the help of
computer vision systems are CMS Mobilizer, Eliop EVA, PEEK
VideoTrak, Nestor Traffic Vision, Sumitomo IDET- and these
systems used region based tracking.
For different automated incident detection (AID) related systems
in North America [2], Guidestar in Minnesota, Transtar in Houston,
TX and COMPASS in Torranto, Canada and in all Dallas-ft worth,
TX Area. Europe developed similar things throughout European
union roadways. Econolite and Citilog Companies had suffered
problem with false detection, for AID. A fog detection system
developed in Holland using visibility sensors to monitor different
weather conditions. Some models or say the systems are already
developed in other countries for coordination of bus stops, their
photographs and other parameters, Automated transmits Stop
inventory model (ATSIM) and ESRI Arc-pad [6].
Furthermore good architecture, planning of transportation with
respect to smart city becomes easier for transportation management
services. Keystone Architecture Required for European Networks
(KAREN) Projects, Framework Architecture Made for Europe smart
(FRAME –S), Extending the FRAME architecture (E-FRAME),
European Project COOPerative SystEMS for Intelligent Road Safety
(COOPERS), Cooperative Vehicle Infrastructure Systems (CVIS)
and Cooperative Vehicles and Road Infrastructure for Road Safety
(SAFESPOT). For Preparation for Driving Implementation and
Evaluation of C2X Communication Technology (PREDRIVE C2X)
FP7-integrated project and many more listed by Souﬁene Djahel and
his team, in their research paper [11].
In these research projects are still going on, due to system
problems with traffic congestion, snow, glare, rain, shadow such like
incidents, different lighting conditions makes CCTV work difficult
and many more.
6. CONCLUSION AND FUTURE WORK
In this paper, based on different research paper read in the context
of transportation management service, explain different algorithms
already used, different project and systems working related to TMS,
smart-phone involvement regarding improvement in traffic
monitoring as well as tracking and in other applications also. Most of
the work explained or designed in a foreign country for smart city
and what's actually required as per our Indian government, they are
trying to develop smart city concept step by step. So for that
initiative from the smart city different parameter better
implementation TMS is required. So now from these review papers,
our aspects of future work –on transportation management service
for INDIAN smart cities and give some solution regarding problem
related to traffic management already discussed in section II C.
ACKNOWLEDGMENT
IJSER

ISSN 2229-5518
IJSER © 2017
In this paper, we review TMS service with reference to different
literature available from TRANSPORTAION RESEARCH Part and
different IEEE conference and journal Papers.
REFERENCES
[A] http://smartcities.gov.in/
[B] http://mrunal.org/2015/06/economy-lecture-smart-cities-swachh-
bharat-rurban-infrastructure.html
[1] Benjamin Coifman, David Beymar, Philip McLauchlan,
Jitendra Malik, ”A real time computer vision system for vehicle
tracking and traffic surveillance”, Transporation Research Part
C 6 (1998), 271-288, 1999 Elsevier Science Ltd.
[2] Mohamed S. Shehata, Jun Cai, Wael Maged Badawy, Tyson W.
Burr, Muzamil S. Pervez, Robert J. Johannesson, and Ahmad
Radmanesh, “Video-Based Automatic Incident Detection for
Smart Roads: The Outdoor Environmental Challenges
Regarding False Alarms”, IEEE TRANSACTIONS ON
INTELLIGENT TRANSPORTATION SYSTEMS, VOL. 9,
NO. 2, JUNE 2008.
[3] A. Ando, T. Morikawa, T. Miwa, T. Yamamoto, “Study of the
acceptability of a parking deposit system as an alternative road
pricing scheme”, IET Intell. Transp. Syst., 2010, Vol. 4, Iss. 1,
pp. 61–75, doi: 10.1049/it-its. 2009.0030.
[4] Gongjun Yan, Weiming Yang, Danda B. Rawat, Stephen
Olariu, “Smart Parking: A Secure and Intelligent Parking
System”, Observation from the IEEE IV 2010 Symposium,
Digital Object Identifier 10.1109/MITS.2011.940473, Date of
publication: 6 April 2011, IEEE INTELLIGENT
TRANSPORTATION SYSTEMS MAGAZINE • 18 •
SPRING 2011.
[5] Li-Wen Chen and Ta-Yin Hu, Member, IEEE, “Flow
Equilibrium under Dynamic Traffic Assignment and Signal
Control—an Illustration of Pretimed and Actuated Signal
Control Policies”, IEEE TRANSACTIONS ON
INTELLIGENT TRANSPORTATION SYSTEMS, VOL. 13,
NO. 3, SEPTEMBER 2012.
[6] Marta Santos, Ricardo Lopes Pereira, António Brandão Leal,
“GBUS - Route GeoTracer”, 978-1-4673-5029-7/12/$31.00
©2012 IEEE, DOI: 10.1109/VTM.2012.6398697.
[7] Anutosh Maitra, Saurabh Bhadkaria, Chiranjeeb Ghosh, Sanjoy
Paul, “An Integrated Transport Advisory System for
Commuters, Operators and City Control Centers”, 978-1-
4673-5029-7/12/$31.00 ©2012 IEEE, DOI:
10.1109/VTM.2012.6398699.
[8] Claudia Campolo, Antonio Iera, Antonella Molinaro,
StefanoYuriParatore, Giuseppe Rugger, “SMaRTCaR: An
Integrated Smartphone-basePlatform to Support Traffic
Management Applications”, 978-1-4673-5029-7/12/$31.00
©2012 IEEE, DOI: 10.1109/VTM.2012.6398700.
[9] Vi Tran Ngoc Nha, Soufiene Djahel and John Murphy, “A
Comparative Study of Vehicles’ Routing Algorithms for Route
Planning in Smart Cities”, 978-1-4673-5029-7/12/$31.00
©2012 IEEE, DOI: 10.1109/VTM.2012.6398701.
[10] Rosamaria Elisa Barone, Tullio Giuffrè, Sabato Marco
Siniscalchi, Maria Antonietta Morgano, Giovanni Tesoriere,
“Architecture for parking management in smart cities”, IET
Intell. Transp. Syst., 2014, Vol. 8, Iss. 5, pp. 445–452,doi:
10.1049/iet-its.2013.0045
[11] Soufiene Djahel, Ronan Doolan, Gabriel-Miro Muntean, and
John Murphy, “A Communications-Oriented Perspective on
Traffic Management Systems for Smart Cities: Challenges and
Innovative Approaches”, IEEE COMMUNICATION
SURVEYS & TUTORIALS, VOL. 17, NO. 1, FIRST
QUARTER 2015.
[12] SaId ALABDALLAOUI, Abdelghafour BERRAISSOUL,
“Performance evaluation of RED approach for traffic lights
management”, 978-1-4673-8709-5/15/$31.00 ©20 15 IEEE.
[13] Fenghua Zhu, Member, IEEE, Zhenjiang Li, Songhang Chen,
and Gang Xiong, “Parallel Transportation Management and
Control System and Its Applications in Building Smart Cities”,
IEEE TRANSACTIONS ON INTELLIGENT
TRANSPORTATION SYSTEMS, VOL. 17, NO. 6, JUNE
2016.
[14] Gitae Kim, Yew Soon Ong, Taesu Cheong, and Puay Siew Tan,
“Solving the Dynamic Vehicle Routing Problem Under Traffic
Congestion”, IEEE TRANSACTIONS ON INTELLIGENT
TRANSPORTATION SYSTEMS, VOL. 17, NO. 8, AUGUST
2016.
IJSER

Review Paper on Intelligent Traffic Control system using Computer Vision for Smart City

More Related Content

What's hot (20)

Similar to Review Paper on Intelligent Traffic Control system using Computer Vision for Smart City (20)

Recently uploaded (20)

Review Paper on Intelligent Traffic Control system using Computer Vision for Smart City