Vision based Automatic Parking System

N.Mohanasundaram*, Siddesh G.K** and H.R.Shashidhara***

*Assistant Professor, JSSATE, Noida
Sundar26865@gmail.com
**Associate Professor, JSSATE, Bangalore
siddeshgundagatti@gmail.com
***Assistant Professor, JSSATE, Bangalore
shashidharahr@gmail.com

Abstract: This paper explains the approach to make easy to monitor and manage a parking area using a vision based
automatic parking system. To find the available parking space in the most efficient manner and to avoid traffic congestion in
the parking area, it is very necessity to manage the car parking. Nowadays by human personnel also we do manage, the car
parking area and by using Sensors, we do monitor the availability of car park spaces. Even though in both we get to know the
number of free parking spaces, but we don’t get the actual location available. The installation and maintenance cost of a
Sensor based system varies as per the number of Sensors required. This paper shows that the performance has a more
accuracy for a model of 4 space car park. The model indicated that the application of a vision based car parking management
system would be able to detect and indicate the duration of car parking in the given space.

Keywords: Object classification, Color detection, Car Park, Vision based, region of interest.

Introduction
Parking of cars in a parking area is a very difficult task as the number of cars are getting increased day by day but the number
of parking spaces are a finite one. Due to that, we spend more time in looking for parking space, by slowing the speed of the
vehicle causing a traffic congestion in the parking area. This problem is mainly due to the non availability of the information
about the availability of parking spaces at that time. We are using different systems like [5] wireless Sensor network systems
and a vision based system [6] now-a-days.

Background
Now-a-days, parking a car is a serious problem in big cities having more population of cars [3].Since the normal parking
system is not providing the required information about the available empty parking areas, we need to improve the system
effectively. The typical car parking system is able to give information about the available parking empty spaces or other
system using human effort to get and provide information about the location of the available empty parking spaces. Since
these parking systems are not giving required information about the available parking spaces and not giving solutions
effectively on time, the drivers are spending a more time in locating the parking spaces.
Sensors can be used to detect and provide the information about the location of available parking spaces and to manage the
problem of parking. Wireless Sensor system [5] is one of the solution. But we need to install Sensors for each parking space
for giving the information on the status of each car park locations, so that the cost of installation of Sensors in each parking
slot is very tedious and its installation cost is proportionally high for the more number of parking slots.
A vision based system is an alternative to the normal system to detect available parking slot locations [5, 6, 7]. The
availability of a vision based system would be cheaper or scalable with only with the utilization of a number of cameras. So
in this method of detecting and locating available parking locations using a vision based system is robust and cost effective
since only one camera is needed at the parking space scanning slots one after another.
The main purpose of this paper is to demonstrate the development of a low cost vision based parking system which can
determine the location of available parking slots and to provide feedback on the required parking location.

Feature Extraction
In this paper, the functions of working of the system starts as follows:
 The user reaches the entrance of the car park and gets the information about the empty parking slots available, after
punching the passive RFID Tag having some identification number for each Tag.
 The user parks the car where the slot is empty. After parking, again he punches the passive RFID Tag having made entry
of parked slot, time like details.
 Vision based Automatic Parking System 537

 When the user comes back to retrieve the car back, again he punches the passive RFID Tag and gets the information
about the parked slot.
 The user pays the parking fee as per the duration of the parking.

The Implemented System

In this paper, both Hardware and Software Systems are used for parking access and Payment. The Hardware details are
described clearly in the given block diagram used for the development of prototype of the Vision Based Automatic Parking
System. Each block, is explained in the Hardware section. The overall block diagram is given in the Fig.1 below as follows:

Fig.1 Proposed system Block Diagram

The Hardware
The proposed prototype is composed of different elements of block as shown in Fig.1. Its functions are as follows:
 Web camera is focussed to each parking slot by one at a time. It gives information about the size of the slot and whether it
is empty or not.
 PC or Lap Top
 Microcontroller is interfaced with the PC to get the information about the parking space availability.
 RFID Reader which is interfaced with the controller to read the information in the RFID Tag for each car.
 IR Obstacle Sensor detects the entry of vehicle and gives the information to the controller.
 LCD gives the sequence of events starting from establishing the interface between PC and Microcontroller, Parking Space
details, Parked Slot details etc.,
 Stepper Motor which controls the entrance gate, opens only if the empty slot is available for parking.
 LED’s which is interfaced with the microcontroller indicates whether the respective parking slot is occupied or not.

The Software
Next to the hardware implementation of the prototype, ad-hoc software structure have been studied and developed .The
requirements of the software are:
 Embedded C
 KEIL compiler
 Flash Magic
 Microsoft visual studio 2008
 Dot net frame work
 MS Access
First of all, a data base has been realized, to store all the data concerning the parking slots and the information about the
time’s inside the parking. MS-Access has been employed as a DBMS.
Subsequently the software Dot net computing the Web camera information about the parking slots of the prototype having 4
different slots P1, P2, P3 and P4. P1 and P4 are of bigger size slots whereas P2 and P3 are of smaller ones. Each slot is
marked with different colors at particular area, like P1 with Red, P2 with Violet, P3 with Green and P4 with Blue. When the
Web camera detects different colors, in the parking slots, it indicates bigger slots P1 or P4 is empty. Similarly smaller slots
P2 or P3.On the Dot net output window, we can get details like interface status, Space availability as flag status, Parked slot
details with entry parking time stamp and exit parking time stamp, Parking slot details P1, P2, P3 or P4.
538 International Conference on Signal, Image Processing Communication and Automation - ICSIPCA- 2017

Flow Chart

Fig.2 Flow Chart for display

This prototype in coordination with the functions of hardware and software, operates in a program controlled sequences like
mentioned in the flow chart shown in Fig.2
The Dot net output window shows the different sequences of functions as per the software written. The snapshots of the
output results are shown in the Fig.3. given below

Test Result : For knowing the Parked Slot details

dot net - output

Fig.3 Dot net output window

Conclusion & Future Work

The Conclusion of this prototype has the following comments.
• In this project, the automatic parking of cars are monitored using ARM7TDMI S Microcontroller as the core which has
excellent built-in peripherals and operates at CPU operating voltage range (1.65V-1.95.5V). It is communicated with PC
which is interfaced to web camera module.
• The information of the parking space having different size slots are monitored.
• As soon as the car enters the area, the entrance gate will be open only if the empty space is available for parking.
• Different sizes of parking slots are available according to the requirement based on the size of the incoming cars.
• For the demonstration model, four parking slots are used. First and fourth slots are bigger type whereas the second and
third slots are smaller type.
• The Future Work for this model can be tried with the following comments.
 Vision based Automatic Parking System 539

• For further improvement in this project, another web camera can be fitted at the entrance to capture the actual size of the
incoming cars, so that the availability of the parking slots can be effectively verified for full satisfaction.

Acknowledgement
The authors would like to thank the management of JSSATE-Noida for giving permission for the presentation & sponsoring
for the registration of this paper.

References
[1] Hamada R.H.Al-Absi, Patrick Sebastian, Justin Dinesh Daniel Devaraj , Yap Vooi Voon, ” Vision-Based Automated Parking
System ,” in Proc. Of 10th International Conference on Information Science,Signal Processing and their Applications(ISSPA 2010) ,
2010 IEEE.
[2] Farhan Manzoor and Karsten Menzel, University College Cork (UCC),Ireland,” Indoor Localisation for Complex Building Designs
using Passive RFID Technology,” 2011 IEEE.
[3] Giuliano Benelli,Alessandro Pozzebon University of Siena,Italy “ An Automated Payment System for Car Parks Based on New Field
Communication Technology”,
[4] Marcelo Zottolo, Kathryn Peacock, Eric Lammers, Edward Williams , PMC , MI 48126, USA.” THROUGHPUT CAPACITY
VERIFICATION OF AUTOMATED PARKING SYSTEMS,” 2008 IEEE.
[5] V.W.S.Tang, Y.Zheng , and J.Cao, “An Intelligent Car Park Management System based on Wireless Sensor Networks,” Proceedings
of the 1st International Symposium on Pervasive Computing and Applications, pp.65-70,Aug 2006.
[6] Sheng-Fuu Lin , Yung-Yao Chen, “A vision based parking lot management system ,” 2006 IEEE Conference on Systems, Man ,
and Cybernetics.Peter. C, pp. 2897-902, Oct.2006.
[7] Wu,C.Huang,S.yuWang,W.chen Chiu and T.Chen. “ Robust parking space detection considering interspace correlation,” In
Proceedings of IEEE International Conference on Multimedia and Expo, pages 659-662,2007.
[8] The 13th IEEE International Symposium on Consumer Electronics (ISCE 2009).
[9] N.True .Vacant parking space detection in static images. University of California , San Diego,2007
[10] Philip Ian Wilson, John Fernandez, “ Facial feature detection using Haar classifiers”, Journal of Computing Sciences in
Colleges,v.21 n.4,p.127-133, April 2006.
[11] Rainer Lienhart and Jochen Maydt, An Extended Set of HAAR-like Features for Rapid Object Detection. IEEE ICIP 2002 , Vol. 1,
pp. 900-903,Sep.2002.
[12] Andrew N.SLOSS, Dominic SYMES, Chris WRIGHT, John Rayfield, “ ARM System Developer’s Guide – Designing and
Optimizing System Software ,” . MORGAN KAUFMANN PUBLISHERS ,2004.
[13] Philips A. Laplante, ” Real time Systems Design & Analysis”., John Wiley & Sons INC publication , 2004
[14] LPC 2103 – Users Manual UM10161, Volume 1, Rev.01-11,January 2006, Philips Semiconductors.
[15] Muhammad Ali Mazadi and Janice Gillespie Mazadi and Rollin D. Mckinlay, “ The 8051 microcontrollers and Embedded Systems
using assembly and c ” , PHI, 2006 / Pearson 2006.
[16] Kirk Zurell “C programming for embedded Systems”, Publishers Group west 1700 Fourth Street, Berkeley, CA 94710, ISBN 1-
929629-04-4
[17] Daniel W Lewis “Fundamentals of Embedded Software Where C and Assembly meet”, Publisher Prentice Hall Inc, 2002, ISBN 0-
471-22855-9.
[18] Dr. K.V.K.K. Prasad, “Embedded Real Time Systems” , Dreamtech Press Publication, 2nd Edition, 2005.