Content uploaded by Praveenraj Pattar
Author content
All content in this area was uploaded by Praveenraj Pattar on Nov 27, 2015
Content may be subject to copyright.
IEEE SENSORS JOURNAL, VOL. 15, NO. 8, AUGUST 2015 4313
Automatic Detection and Notification of Potholes
and Humps on Roads to Aid Drivers
Rajeshwari Madli, Santosh Hebbar, Praveenraj Pattar, and Varaprasad Golla
Abstract—One of the major problems in developing countries
is maintenance of roads. Well maintained roads contribute a
major portion to the country’s economy. Identification of pave-
ment distress such as potholes and humps not only helps drivers
to avoid accidents or vehicle damages, but also helps authorities to
maintain roads. This paper discusses previous pothole detection
methods that have been developed and proposes a cost-effective
solution to identify the potholes and humps on roads and provide
timely alerts to drivers to avoid accidents or vehicle damages.
Ultrasonic sensors are used to identify the potholes and humps
and also to measure their depth and height, respectively. The
proposed system captures the geographical location coordinates
of the potholes and humps using a global positioning system
receiver. The sensed-data includes pothole depth, height of hump,
and geographic location, which is stored in the database (cloud).
This serves as a valuable source of information to the government
authorities and vehicle drivers. An android application is used
to alert drivers so that precautionary measures can be taken to
evade accidents. Alerts are given in the form of a flash messages
with an audio beep.
Index Terms—Android application, GSM SIM900, GPS,
PIC16F877A, ultrasonic sensors.
I. INTRODUCTION
I
NDIA, the second most populous Country in the World
and a fast growing economy, is known to have a
gigantic network of roads. Roads are the dominant means of
transportation in India today. They carry almost 90 percent of
country’s passenger traffic and 65 percent of its freight [1].
However, most of the roads in India are narrow and congested
with poor surface quality and road maintenance needs are not
satisfactorily met. No matter where you are in India, driving
is a breath-holding, multi-mirror involving, potentially life
threatening affair.
Over the last two decades, there has been a tremendous
increase in the vehicle population. This proliferation of
vehicles has led to problems such as traffic congestion and
increase in the number of road accidents. Pathetic condition of
roads is a boosting factor for traffic congestion and accidents.
Researchers are working in the area of traffic congestion
Manuscript received December 13, 2014; revised February 4, 2015 and
March 18, 2015; accepted March 18, 2015. Date of publication March 30,
2015; date of current version June 10, 2015. This work was supported
by the All India Council for Technical Education. The associate editor
coordinating the review of this paper and approving it for publication was
Prof. Subhas C. Mukhopadhyay.
The authors are with the Department of Computer Science and
Engineering, B.M.S. College of Engineering, Bangalore 560019, India
(e-mail: m.rajeshwari1626@gmail.com; santhoshshebbar@gmail.com;
pattar.praveenraj3537@gmail.com; drvaraprasad@gmail.com).
Color versions of one or more of the figures in this paper are available
online at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/JSEN.2015.2417579
Fig. 1. Condition of roads with potholes.
control [2], an integral part of vehicular area networks, which
is the need of the hour today.
Roads in India normally have speed breakers so that the
vehicle’s speed can be controlled to avoid accidents. However,
these speed breakers are unevenly distributed with uneven and
unscientific heights.
Potholes, formed due to heavy rains and movement of heavy
vehicles, also become a major reason for traumatic accidents
and loss of human lives. According to the survey report “Road
Accidents in India, 2011”, by the ministry of road transport
and highways, a total of 1,42,485 people had lost their lives
due to fatal road accidents. Of these, nearly 1.5 per cent or
nearly 2,200 fatalities were due to poor condition of roads.
Figure 1 portrays the condition of roads with killer potholes.
To address the above mentioned problems, a cost effective
solution is needed that collects the information about the
severity of potholes and humps and also helps drivers to drive
safely. With the proposed system an attempt has been made
to endorse drivers to ward off the accidents caused due to
potholes and raised humps.
The remaining sections of the paper are as follows:
section II emphasises on the related work that has been done
and is going on in the field of detection of potholes and
humps. Section III discusses the various components used in
the proposed system. Section IV describes the architecture and
implementation of the proposed system. Experimental results
of the proposed work are presented in Section V. Section VI
talks about conclusion and future scope.
II. R
ELATED WORK
Pavement distress detection is an intriguing topic of
research and researchers have been working on pothole
1530-437X © 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
4314 IEEE SENSORS JOURNAL, VOL. 15, NO. 8, AUGUST 2015
detection techniques. This section gives a brief description
about the existing solutions for detecting potholes and humps
on roads.
Moazzam et al. [3] have proposed a low cost model for
analysing 3D pavement distress images. It makes use of a low
cost Kinect sensor, which gives the direct depth measurements,
thereby reducing computing costs.
The Kinect sensor consists of a RGB camera and an
IR camera, and these cameras capture RGB images and depth
images. These images are analysed using MATLAB environ-
ment, by extracting metrological and characteristic features, to
determine the depth of potholes. Youquan et al. [5] developed
a model to detect the three-dimensional cross section of
pavement pothole. The method makes use of LED linear
light and two CCD (Charge Coupled Device) cameras to
capture pavement image. It then employs various digital image
processing technologies including image pre-processing,
binarization, thinning, three dimensional reconstruction, error
analysis and compensation to get the depth of potholes.
However, results get affected by LED light intensity and envi-
ronmental factors. Lin and Liu [6], have proposed a method for
pothole detection based on SVM (Support Vector Machine).
This method distinguishes potholes from other defects such as
cracks. The images are segmented by using partial differential
equations. In order to detect potholes, the method trains the
SVM with a set of pavement images. However, the training
model fails to detect the pavement defects if the images are
not properly illuminated. Orhan and Eren [7], have proposed
a work developed on android platform to detect road hazards.
There are three components in this proposed work viz, Sensing
component, Analysis component and Sharing component. The
sensing component basically works by collecting raw data
from accelerometer and synchronizes with interface, hence
leading to ease of access. In analysis component, the values
obtained from the sensors are used for developing analysis
modules. The sharing component works as follows: the
developed framework is connected with the central application,
where it can directly communicate with the social network.
All the collected data is stored at central repository for further
processing. Although this method communicates traffic events
with other drivers, it increases the cost and complexity of
implementation.
Mednis et al. [8] have proposed a real time pothole detection
model using Android smartphones with accelerometers.
Modern smart phones with android OS, have inbuilt
accelerometers, which sense the movement and vibrations.
The accelerometer data is used to detect potholes. Different
algorithms such as Z-thresh, which measures the acceleration
amplitude at Z-axis, Z-diff to measure the difference between
the two amplitude values, STDEV (Z) to find the standard
deviation of vertical axis acceleration and G-Zero are used
to identify potholes. Zhang et al. [9] have made use of stereo
camera images coupled with a disparity calculation algorithm
to identify potholes. The location coordinates of the potholes
are also captured and stored in the database. Strutu et al. [10]
have proposed a method for detecting defects on the road
surface using accelerometers. It also makes use of GPS system
to identify the exact location of the defects. Pothole detection
algorithm runs on a mobile platform (moving vehicles),
which is installed with accelerometer, GPS, local computer
and a wireless router. The sensed data is communicated to the
central database using primary access points and secondary
access points which can be used for future processing.
However, installing wireless router and local computer on
all mobile platforms and setting up access points turns out
to be quite expensive. Murthy and Varaprasad et al. [11],
have proposed a system that detects potholes based on a
vision based approach. The pictures of the road surface are
captured using a properly mounted camera. The images are
then processed using MATLAB to detect the occurrence
of potholes. It is a 2D vision based solution and works
only under uniform lighting conditions and also the system
does not involve any kind of warning system. The above
solutions are limited only to the identification of a pothole.
These solutions do not provide any aid to the driver to avoid
accidents due to potholes and humps.
Rode et al. [4], have proposed a system in which, Wi-Fi
equipped vehicles collect information about the road surface
and pass it to the Wi-Fi access point. The access point then
broadcasts this information to other vehicles in the vicinity in
the form of warnings. However, the system turns out to be an
expensive one as all vehicles should be installed with Wi-Fi
stations and more number of access points have to be set up.
Venkatesh et al. [12] have proposed an intelligent system that
has made use of laser line striper and a camera to detect and
avoid potholes. This system maintains a centralized database
of the location of potholes. It also sends warning messages
to the nearby vehicles about the occurrence of potholes
using Dedicated Short Range Communication protocol.
Hegde et al. [13], have proposed an intelligent transport system
to detect potholes. It makes use of ultrasonic sensors to detect
the presence of potholes. This system also sends warning
messages to all the vehicles in the range of 100 meters using
Zigbee module. However, the system provides warnings after
detecting the potholes which does not effectively help drivers
to avoid potential accidents.
More et al. [15], proposed a system where sensors are
mounted on public vehicles. These sensors record vertical and
horizontal accelerations experienced by vehicles on their route.
The installed GPS device logs its corresponding coordinates
to locate potholes and the collected data is processed to locate
potholes along the path traversed earlier by the vehicle. A Fire
Bird V robot is used for experimenting with constant speed.
The moving robot is mounted with a servo motor which
rotates 0-180 degrees along with IR Sharp sensors. IR Sharp
sensors check for variance in constant speed. If variance is
detected, it is an indication of a pothole; robot stops and
camera moves to take pictures of the pothole while GPS
device locates its coordinates. Although this is a cost effective
solution, it is restricted to collecting information about
potholes. Yu and Salari [16], implemented a system that uses
laser imaging for detecting potholes. Pavement distress such
as pothole is detected when the laser source deformation is
observed in the captured images. Different techniques such as
multi-window median filtering and tile partitioning are applied
to detect the presence of potholes. These potholes are further
MADLI et al.: AUTOMATIC DETECTION AND NOTIFICATION OF POTHOLES AND HUMPS ON ROADS TO AID DRIVERS 4315
Fig. 2. Working principle of ultrasonic sensor.
classified based on their shapes and severity. Although this is
an accurate and efficient method for detecting potholes, the
cameras capture shaky images due to uneven road surface,
which reduces the efficiency of pothole detection.
Chen et al. [17] proposed a system for detecting potholes
using GPS sensor and three-axis accelerometer. The outputs
are taken from the GPS sensor and three-axis accelerometer
and fed into data cleaning algorithm. In the second part of
the implementation the inputs to the algorithm are processed
for power spectra density (PSD) to calculate the roughness
of potholes. After analysing, roughness is classified into
different levels.
III. C
OMPONENTS USED IN THE PROPOSED SYSTEM
The proposed system offers a cost effective solution for
detecting potholes and humps on roads and notifying drivers
about their presence. Components used in the proposed work
are as follows:
PIC 16F877A Microcontroller: Peripheral Interface Control
(PIC 16F887A) is a 40 pin microcontroller with 8k program
memory. It is widely used due to its low cost, high application
support and wide availability. Microcontroller is the heart of
the proposed system and is responsible for performing various
tasks starting form processing all the sensor inputs to alerting
the driver.
Ultrasonic Sensors HC-SR04: The HC-SR04 is an active
ultrasonic sensor and contains a transmitter and a receiver. It is
used to measure distance at which, objects are placed in front
of it. The ultrasonic sensor transmits high frequency sound
waves and waits for the reflected wave to hit the receiver. The
distance is calculated based on the time taken by the ultrasonic
pulse to travel a particular distance [19]. The working principle
of this device is shown in figure 2. There are different types
of ultrasonic sensors with different transmission ranges and
angles of detection. The HC-SR04 sensor work at frequency
of 40 KHz and can measure distances of the objects in the
range 2 to 400 cm with a 15° angle of detection.
GPS Receiver: Global Positioning System (GPS) is a
satellite navigation system and is used to capture geographic
location and time, irrespective of the weather conditions. It is
maintained by the U.S. Government and is freely available
to anyone who has a GPS receiver. It obtains the GPS
information from satellites in National Marine Electronics
Association (NMEA) format. The NMEA has defined a
standard format for the GPS information. This is followed by
all the satellites. The standard defines various codes such as
Fig. 3. Architecture of the proposed system.
GLL-Latitude/Longitude data, GSV–Detailed satellite data and
RMC-Minimum Recommended Data [14].
GSM SIM 900: Global Standards for Mobile
Communication (GSM) is a set of standards for Second
Generation (2G) cellular networks. The GSM SIM 900 module
uses any network provider’s SIM to communicate over the
telecommunication network. This modem can be used to
send and receive text messages and to make and receive
voice calls. GSM SIM 900 is a quad-band GSM modem that
functions at 850, 900, 1800 and 1900 MHz frequencies. This
modem also supports features like transferring voice data,
integrated support for GPRS and TCP/IP stack.
IV. A
RCHITECTURE &IMPLEMENTATION
The architecture of the proposed system is shown in figure 3.
It consists of 3 parts; microcontroller module, server module
and the mobile application module. Microcontroller module
is used to gather information about potholes and humps
and their geographical locations and this information is sent
to the server. Server module receives information from the
microcontroller module, processes and stores in the database.
Mobile application module uses information stored in the
server database and provides timely alerts to the driver.
Microcontroller Module: This module consists of
4 components, namely, PIC 16F877A microcontroller,
ultrasonic sensors, GPS receiver and GSM modem. Ultrasonic
sensors are used to measure the distance between the car
body and the road surface and this data is received by
the microcontroller. The distance between car body and
the ground, on a smooth road surface, is the threshold
distance. Threshold value depends on the ground clearance
of vehicles and can be configured accordingly. If the distance
measured by ultrasonic sensor is greater than the threshold,
it is a pothole, if it is smaller, it is a hump otherwise it is
a smooth road.
The GPS receiver captures the location coordinates of the
detected pothole or the hump and sends messages to the
4316 IEEE SENSORS JOURNAL, VOL. 15, NO. 8, AUGUST 2015
Fig. 4. Workflow of mobile application.
registered mobile SIM using GSM modem. This registered
mobile SIM is present on the android device that acts as server.
The messages sent include information about depth of the
pothole or height of the hump and its location coordinates.
Server Module: This module consists of two parts; the
android device and the database. It acts as an intermediary
layer between the microcontroller module and the mobile
application. The server module is implemented as an android
application that runs on a device and is responsible for
reading messages sent by the registered mobile SIM present
in the microcontroller module. It processes the contents of
this message and stores it in the database (cloud). Integrating
sensor networks with cloud and Internet of Things [18], it is
possible to allow broader access to sensor data.
Mobile Application Module: This module is implemented as
an android application that is installed on the vehicle driver’s
mobile phone to provide timely alerts about the presence
of potholes and humps. Figure 4 shows the workflow of
this application. The application continuously runs in the
phone background. It first captures the current geographic
location of the vehicle and then accesses the locations of
potholes and humps stored in the server database. The distance
between the vehicle location and the pothole location stored
in database is computed. If the distance between the two is
within 100 meters, an alert message pops up on the mobile
screen. This message is accompanied with an audio beep so
that the driver can differentiate it from other flash messages.
V. E
XPERIMENTAL RESULTS
The working model of the proposed system is shown
in figure 5. It was tested in a simulated environment with
artificial potholes and humps. The model was also tested
in real time by fixing it on a motor bike (Honda Activa).
Fig. 5. Working model of the proposed system.
TABLE I
I
NFORMATION ABOUT POTHOLES AND HUMPS COLLECTED
IN
SIMULATED TEST ENVIRONMENT
Fig. 6. Message sent to android device about the pothole and hump locations.
Tests were carried out in two phases. In the first phase,
information about potholes and humps was recorded and
stored in the server database. In second phase, alerts were
generated based on pothole and hump information stored
in database. While testing in the simulated environment,
MADLI et al.: AUTOMATIC DETECTION AND NOTIFICATION OF POTHOLES AND HUMPS ON ROADS TO AID DRIVERS 4317
Fig. 7. Pothole alert displayed on the mobile phone.
Fig. 8. (a) Proposed model fixed on two wheeler bike for testing.
(b) Detection of lump. (c) Detection of pothole.
the microcontroller module was fixed on a toy-car and the
threshold value was configured to 5 cm. During the tests
it was found that the microcontroller module worked as
expected to identify potholes and humps. Table I shows a
set of potholes and humps identified by the system in the
simulated environment. Information about potholes and humps
was successfully sent to the android device (server). The
snapshot of these messages can be seen in figure 6. The server
processed the messages received and stored in the database.
In the above table, obstacle type ‘P’ indicates a pothole
and ‘H’ indicates a hump.
In the second phase of testing, the mobile application that
generates alerts was successfully tested by moving the toy-car
on routes containing potholes and humps and alerts were
generated for potholes and humps recorded in the first phase.
Figure 7 shows an alert generated by this application.
Figure 8 shows the real time testing of the proposed model.
The microcontroller module was fixed on Honda Activa and
the threshold distance value was configured to 16 cm, which
is the ground clearance for Honda Activa. The vehicle was
TABLE II
I
NFORMATION ABOUT POTHOLES AND HUMPS COLLECTED
DURING REAL-TIME TESTING
moved on Bangalore roads for the purpose of recording
information about potholes and humps, and the test results
were as expected. Table II shows a set of potholes and humps
detected during real-time tests.
VI. C
ONCLUSION AND FUTURE RESEARCH WORK
The model proposed in this paper serves 2 important
purposes; automatic detection of potholes and humps and
alerting vehicle drivers to evade potential accidents. The
proposed approach is an economic solution for detection
of dreadful potholes and uneven humps, as it uses low
cost ultrasonic sensors. The mobile application used in this
system is an additional advantage as it provides timely
alerts about potholes and humps. The solution also works
in rainy season when potholes are filled with muddy water
as alerts are generated using the information stored in the
database. We feel that the solution provided in this paper can
save many lives and ailing patients who suffer from tragic
accidents.
The proposed system considers the presence of potholes and
humps. However, it does not consider the fact that potholes or
humps get repaired by concerned authorities periodically. This
system can be further improved to consider the above fact and
update server database accordingly. Also, Google maps and
SATNAV can be integrated in the proposed system to improve
user experience.
R
EFERENCES
[1] India Transport Sector. [Online]. Available: http://web.
worldbank.org/WBSITE/EXTERNAL/COUNTRIES/SOUTHASIAEXT/
EXTSARREGTOPTRANSPORT/0„contentMDK:20703625~menuPK:
868822~pagePK:34004173~piPK:34003707~theSitePK:579598,00.html,
accessed Mar. 16, 2015.
[2] R. Sundar, S. Hebbar, and V. Golla, “Implementing intelligent traffic
control system for congestion control, ambulance clearance, and stolen
vehicle detection,” IEEE Sensors J., vol. 15, no. 2, pp. 1109–1113,
Feb. 2015.
[3] I. Moazzam, K. Kamal, S. Mathavan, S. Usman, and M. Rahman,
“Metrology and visualization of potholes using the microsoft Kinect
sensor,” in Proc. 16th Int. IEEE Conf. Intell. Transp. Syst., Oct. 2013,
pp. 1284–1291.
[4]S.S.Rode,S.Vijay,P.Goyal,P.Kulkarni, and K. Arya, “Pothole
detection and warning system: Infrastructure support and system
design,” in Proc. Int. Conf. Electron. Comput. Technol., Feb. 2009,
pp. 286–290.
[5] H. Youquan, W. Jian, Q. Hanxing, Z. Wei, and X. Jianfang, “A research
of pavement potholes detection based on three-dimensional projection
transformation,” in Proc. 4th Int. Congr. Image Signal Process. (CISP),
Oct. 2011, pp. 1805–1808.
[6] J. Lin and Y. Liu, “Potholes detection based on SVM in the pavement
distress image,” in Proc. 9th Int. Symp. Distrib. Comput. Appl. Bus. Eng.
Sci., Aug. 2010, pp. 544–547.
4318 IEEE SENSORS JOURNAL, VOL. 15, NO. 8, AUGUST 2015
[7] F. Orhan and P. E. Eren, “Road hazard detection and sharing with
multimodal sensor analysis on smartphones,” in Proc. 7th Int. Conf.
Next Generat. Mobile Apps, Services Technol., Sep. 2013, pp. 56–61.
[8] A. Mednis, G. Strazdins, R. Zviedris, G. Kanonirs, and L. Selavo, “Real
time pothole detection using Android smartphones with accelerometers,”
in Proc. Int. Conf. Distrib. Comput. Sensor Syst. Workshops, Jun. 2011,
pp. 1–6.
[9] Z. Zhang, X. Ai, C. K. Chan, and N. Dahnoun, “An efficient algorithm
for pothole detection using stereo vision,” in Proc. IEEE Int. Conf.
Acoust., Speech Signal Process., May 2014, pp. 564–568.
[10] M. Strutu, G. Stamatescu, and D. Popescu, “A mobile sensor network
based road surface monitoring system,” in Proc. 17th Int. Conf. Syst.
Theory, Control Comput. (ICSTCC), Oct. 2013, pp. 630–634.
[11] S. B. S. Murthy and G. Varaprasad, “Detection of potholes in
autonomous vehicle,” IET Intell. Transp. Syst., vol. 8, no. 6,
pp. 543–549, Sep. 2013.
[12] S. Venkatesh, E. Abhiram, S. Rajarajeswari, K. M. Sunil Kumar,
S. Balakuntala, and N. Jagadish, “An intelligent system to detect, avoid
and maintain potholes: A graph theoretic approach,” in Proc. 7th Int.
Conf. Mobile Comput. Ubiquitous Netw., 2014, p. 80.
[13] S. Hegde, H. V. Mekali, and G. Varaprasad, “Pothole detection and inter
vehicular communication” in Proc. IEEE Int. Conf. Vehicular Electron.
Safety (ICVES), 2014, pp. 84–87.
[14] GPS. NMEA Data. [Online]. Available: http://www.gpsinformation.org/
dale/nmea.htm, accessed Oct. 19, 2014.
[15] P. More, S. Surendran, S. Mahajan, and S. K. Dubey, “Potholes and
pitfalls spotter,” IMPACT, Int. J. Res. Eng. Technol., vol. 2, no. 4,
pp. 69–74, Apr. 2014.
[16] X. Yu and E. Salari, “Pavement pothole detection and severity measure-
ment using laser imaging,” in Proc. IEEE Int. Conf. EIT, May 2014,
pp. 1–5.
[17] K. Chen, M. Lu, X. Fan, M. Wei, and J. Wu, “Road condition monitoring
using on-board three-axis accelerometer and GPS sensor,” in Proc. Int.
ICST Conf. Commun. Netw. China, Aug. 2011, pp. 1032–1037.
[18] F. Li and P. Xiong, “Practical secure communication for integrating
wireless sensor networks into the Internet of Things,” IEEE Sensors J.,
vol. 13, no. 10, pp. 3677–3684, Oct. 2013.
[19] A. Carullo and M. Parvis, “An ultrasonic sensor for distance mea-
surement in automotive applications,” IEEE Sensors J., vol. 1, no. 2,
pp. 143–147, Aug. 2001.
Rajeshwari Madli received the B.E. degree in computer science and
engineering from Visvesvaraya Technological University, Belgaum, India,
in 2007. She is currently pursuing the M.Tech. degree in computer science
and engineering at the B.M.S. College of Engineering, Bangalore, India. Her
areas of research are vehicular area networks and sensor networks.
Santosh Hebbar is currently pursuing the B.E. degree in electrical and
electronics engineering at the B.M.S. College of Engineering, Bangalore,
India. His areas of research are mobile ad hoc network and sensor networks.
Praveenraj Pattar received the B.E. degree in computer science and engi-
neering from Bangalore University, Bangalore, India, in 2012. He is currently
pursuing the M.Tech. degree in computer science and engineering at the
B.M.S. College of Engineering, Bangalore. His areas of research are vehicular
area networks and sensor networks.
Varaprasad Golla received the B.Tech. degree in computer science and
engineering from Sri Venkateswara University, Tirupati, India, in 1999, the
M.Tech. degree in computer science and engineering from the B.M.S. College
of Engineering, Bangalore, India, in 2001, and the Ph.D. degree in
computer networks from Anna University, Chennai, India, in 2004. He was a
Post-Doctoral Fellow with the Indian Institute of Science, Bangalore,
in 2005. He is currently an Associate Professor with the B.M.S. College of
Engineering. His current research interests include routing algorithms, mobile
communications, and simple network management protocol (SNMP).