IRJET- Smart Parking System using Facial Recognition, Optical Character Recognition and Internet of Things (IoT)

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 04 | Apr 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 1278
SMART PARKING SYSTEM USING FACIAL RECOGNITION, OPTICAL
CHARACTER RECOGNITION AND INTERNET OF THINGS (IoT)
Mayank Swaraj1, Manoj Kumar Munagala2, Prerna Bharti3, Mrs. C. Jayavarthini4
1,2Student, Computer Science Engineering, SRM Institute of Science and Technology, Chennai, Tamil Nadu, India
3
4Assistant Professor, Computer Science Engineering, SRM Institute of Science and Technology, Chennai, India
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract - The regular issue in our urban communities is the
trouble of discovering free parking openings. The parking
issue causes wastage of time and energy especially for those
who go to work and are searching for a place to park. The
main objective is to develop an efficient model to deal with
difficulties, such as improper parking, clog in parking areas,
insecurity of vehicles, etc. A platform is given as entryway for
booking the parking spots. The proposed framework uses
facial recognition and optical character recognition along
with IoT (Internet of Things) for accessing the entrywayofthe
parking spot. The proposed framework diminishes the
searching time of free space. The goal is to limit seeking time,
fuel utilization and CO2 emission and make the process of
parking convenient and hassle- free.
Key Words: Cloud, Database, Face Recognition, Google
Cloud, HTTP, IoT,OpticalCharacterRecognition,Python,
Raspberry Pi, requests, Sensor, Smart Parking System,
SQL
1. Introduction
The population of the world is growingatanalarmingrate.It
is one of the most rising global concerns and if not properly
dealt with would cause a huge amount of problems. Traffic
congestion is one of them. It often leads to wastage of time
and fuel and often makes us irritated.Urbanizationhasledto
decrease in empty spaces available and has also led to
increase in number of vehicles. It is now a lot easier to owna
vehicle than few years ago. Finding a parking spot has now
become a daily activity for many people around the globe. It
may be in office, shopping malls or in colleges. This pursuit
consumes around large amountoftheworld'salreadyscarce
oil each day.
These problems will worsen if not dealt with properly. So
our main objective is to create a smart parking system using
Facial
Recognition and Optical Character Recognition integrated
with an IoT device so that we avoid any misuse of space and
increase the efficiency of parking. We will use Facial
Recognition to identify the driver and Optical character
recognition to identify and extract text in the number plate
of the vehicle. We will have a Database from which we will
verify the driver and the vehicle. Thecarwill beallowedonly
if the driver and the vehicle image matches with that in the
database and IoT would be integrated with the gate of the
parking spot and there will be RFID sensor to detect if the
parking spot is occupied or not.
Source- https://www.happiestminds.com/
Optical Character Recognition ismainlyusedforreadingtext
from images and main application of Face
Recognition is to identify and recognize a face of a person.
Optical Character Recognition (OCR)istheidentificationand
validation of printed or written text characters from an
image. OCR can either be done offline or online. In Offline
Recognition the source is a filtered picture or on the other
hand record while in the case of online recognition the
source is progressive and is represented as a function
dependent on time. In Optical character recognition, the
vehicle number plate image is captured as a picture and
identifies number plate and which is at that point changed
over into text which acts as an input for the database to be
validated. The OCR works on images to identify the vehicle
by the number plate. It analyzes light and dark areas in a
picture to identify text. It looks line by line to find if black
and white dots represent a particular character. The steps
involved in the whole process in the given order are
scanning, location segmentation, pre-processing, feature
extraction and final display of output.
Student, Computer Science Engineering, Institute of Technical Education and Research(ITER), Bhubaneshwar, India

Sample Image for Optical Character Recognition (Source-
Google)
Face Recognition has its fair share of advantages over other
biometric systems such as fingerprint/palm/iris.Distanceis
not a problem in face recognition so identification does not
require direct contact with the person. So time required for
biometric verification will minimize to a great extent which
makes it the most convenient option. There are various
algorithms for facial recognition available. Initially,wehave
images of people at different angles in our database. So the
degree at which car comes would not affect the validation
system.
Sample Image for Face Recognition (Source- Google)
The main concept of Internet of Things (IoT) is that it acts as
communication devices which are able to interact with each
other. The users have the ability to track, control or screen
these devices utilizing remote PCs connected through
Internet gateway. IoT increases utilization of the Internet
giving communication. In IoT nodes or devicescanbesaidto
speak with one another and the environment utilizing the
internet protocols. IoT has begun the start of a new era
where everything can be automated. Creating a SmartCityis
currently getting to be conceivable with the rapid
development in the field of Internet of Things. One of the
major issue that savvy urban communities identify with are
vehicle leaving offices and traffic the board frameworks.
2. Literature Survey
Various ideas have been proposed using various algorithms.
We aim to improve over the negative points in some papers
to create an efficient system for parking. There are various
algorithms especially for facial recognition and Optical
Character Recognition.
In [1] the observations with results are of not more than 2-
second performative action and are 85% accurate. We want
to reduce the time and increase the accuracy and efficiency
of the system.
In [2] we get a good outcome but the variation of image at
different is an issue. So for the proposed facial recognition
algorithm, we want to overcome this problem.
In [3] It takes lot of time to scan, thus making it inconvenient
and the user impatient.
In [4] Results have shown success up to a maximum of 90%
accuracy, regardless of the facial pose variation. The only
issue is with capturing image from only limited angles to
detect vehicles only from frontal and side views.
In [5] Advancements in various technologies hasleadtocost
effective solutions.
In [6] Internet of Things, in combination with the usage of
Ultrasonic sensor has leadtosuccessful resultsbutcannot be
implemented because of high costs.
In [7] the results are satisfactory with a high accuracy of
91%. Its accurate in localizing and recognizing the number
plate with help of the algorithm used.
In [8] RFID and OCR are the technologies being used. It will
be an efficient system from small to large organizationswith
an accuracy of 89%. Using this method in large
organizations will lead to wastage of time as well as makes
the whole process complicated.
In [9] Tracking in a huge scenario can be complicatedaswell
as hectic. Accuracy was found out to be 98%.
In [10] The accuracy was found to be 94% but
implementation was too expensive.
In [11] a wide range of improvement is suggested in the
already existing cloud-based parking systems. The paper
introduces algorithms toincreasetheaccuracy.Theresearch
is based on wireless access in an open-source physical
computing platform in cloud based on Arduino with RFID
technology.
In [12] a Parking System is designed which enables the user
to search for the nearest parking spot. It focuses on

considerably reducing the vital time wasted in finding the
parking spots. Thus, it reduces the fuel consumption and
saves from pollution.
In [13] users are shown which parking slots are free and in
which area. This project is only developed to check the
availability of free parking slots.
Source - https://www.researchgate.net/figure/Trend-of-
smart-parking-porotypes-in-academia_fig1_311535831
Some of the above mentioned papers are very expensive to
be practically used. So we mainly aim to reduce the costs
thus making it possible to implement inthedailyusesystem.
3. Proposed System
The Proposed System is using cameras to scan both the
vehicle and the face. The vehicle number plate is scanned
with the help of Optical Character Recognition Algorithm
whereas face is scanned through Face Recognition
Algorithm. IoT device is used to access the gate of a parking
spot. When we design the database, we will initiallyhavethe
images of each person at different angles just as shown
below so that picture taken from any angle won’t affect the
result. This way safety is increased as no intruders are
allowed. An alarm indication is present if wrongly parked or
someone else intrudes in. The proposed system ensuresthat
the whole process is automated and that the accuracy is not
compromised upon time.
Advantages –
Optimize parking space usage.
Reduces wastage of time.
Enhanced security.
Reduces CO2 emission.
Reduces fuel consumption.
Decreases the chances of car – theft.
Disadvantages –
The Camera is always active even when there are no
vehicles.
It takes quite a budget to set up the actual model.
Technologies Used –
Number Plate Recognition using Optical Character
Recognition.
Facial Recognition System.
Data Processing and Validation from Database.
IR Sensors integrated with Raspberry Pi.
Applications-
Safety is ensured as no intruders are allowed. IoT device
integration assures that the system does not require any
manual work. It can be used for many purposes like:
Parking Areas in Industries: Reserved parking spots for
specific people with higher posts.
Parking areas in colleges/schools:Separateparkingspotsfor
teachers and students. Some reserved spaces for specific
teachers, etc.
Parking areas in Malls/Stadiums.
4. Working Process
First, We Capture the image of the VehicleandtheDriver, we
get the text in the number plate using Optical Character
Recognition and we get the face using Face Recognition. The
driver is stopped at the entry gate, and we capture his/her
photo. We verify from the database, if the vehicle is
registered and if the car owner and the number plate
matches. Then IoT device is used to open the gate. Once the
user enters the correct OTP received on his/her registered
mobile/e-mail the welcome message will be displayed and
the gate is opened for the user to park the vehicle. When the
vehicle is parked in the available slot the binary value of the
IR sensor will change and it will demonstrate Slot
unavailable to other users till the vehicle isn't removed.
Some Spots are reserved forspecific people.Ifsomeother
person park in that reserved spot, the alarm is rung until the
person removes his vehicle from the reserved space. This
way we increase security and reduce car-theft. By specific
allocation of places to people of different categorieswecould
decrease the congestionthusreducingtraffic,savingtimeand
fuel. Thus making the whole process of parking convenient.

5. Future Scope
Face recognition and Optical Character recognition system
used till date works pretty well under limited conditions,
although most of the systems work much better with frontal
pictures and constant bright lighting. Almost every present-
day Face Recognition and Optical CharacterRecognitionfails
to a certain extent under rough conditions. Future
generation recognition systems should identify and
recognize individualsandextractcharactersfromthepicture
in real time and even in harsh environments.
It is firmly believed that identification and recognition
systems that are found to be robust in natural conditions, in
the presence of high noise do not give desired results, so we
can't rely upon a single technology, so it becomes an
absolute necessity to combine two or more technology to
come up with an efficient solution.
Technology should be easy to use and user- friendly.
Considering all the requirements, recognition systems that
use face recognition and optical character recognition seem
to be most potential for wide-spread application.
IoT device is the future of technology. In IoT devices can be
made to communicate with each other and the environment
using the internet protocols.
6. Conclusions
There is going to be rapid growth in the sector of facial
recognition and Optical Character Recognition. A lot of
research is being done so as to make the process more
convenient and at the same time accurate, with an ever-
increasing application in the future. Facial Recognition and
OCR has a huge number of applications and it will be
integrated into various systems in upcoming years. They are
one of the most preferred types in biometric domain. It has
some drawbacks and we aim to overcome it. Using facial
recognition and OCR, our main objective is to reduce the
time for the scan and thus making the whole process
convenient. The result obtainedis verifiedfromthedatabase
to prevent intruder entry and exit.
The future of automation lies in the hand of IoT as it helps
device I communicating. Several work isgoingoninthisfield
to automate several tasks.
7. References
1. Raed Abdulla . Automatic Car Park Management System
Using Face and Vehicle Registration Recognition In Asia
Pacific University of Technology and Innovation.
2. Imene TALEB, Mohamed El Amine OUIS, Madani OULD
MAMMAR . Access Control Using Automated Face
Recognition Based on the PCA & LDA algorithms in2014 4th
International Symposium ISKO-Maghreb.
3. Zahid Mahmood, Tauseef Ali, Shahid Khattak, Samee U.
Khan, Laurence T. Yang. Automatic Vehicle Detection and
Driver Identification Framework for Secure Vehicle Parking
in 2015 13th International Conference on Frontiers of
Information Technology.
4. Nazia Bibi, Muhammad Nadeem Majid, Hassan Dawood,
Ping Guo. AutomaticParking SpaceDetectionSystemin2017
2nd International Conference on Multimedia and Image
Processing.
5. Tsung-hua Hsu, Jing-Fu Liu, Pen-NingYu, Wang-ShuanLee
and Jia-Sing . Development of an Automatic Parking System
for Vehicle in IEEEVehiclePowerandPropulsionConference
(VPPC),China.
6. SARTHAK MENDIRATTA, DEBOPAM DEY ,DEEPIKA RANI
SONA . Automatic Car Parking System with Visual Indicator
along with IoT in International Conference on
Microelectronic Devices, Circuits and Systems (ICMDCS).
7. Ridhi Sharma, Kanishka P. Jasani, Ankita R. Jaiswal,
Poonam R. Bais . Automated Parking and Security System in
International Journal of Advance research, ideas and
innovations in technology.
8. M. M. Rashid, A. Musa, M. Ataur Rahman, and N. Farahana,
A. Farhana. Automatic Parking Management System and
Parking Fee Collection Based on Number Plate Recognition
in International Journal ofMachineLearningandComputing.
9. Yadnesh Joshi, Pratik Gharate, Chetan Ahire, Nikhil Alai,
Samadhan Sonavane. Smart Parking Management System
Using RFID and OCR in 2015 International Conference on
Energy Systems and Applications.
10. Kartikeya Jain, Tanupriya Choudhury, Nirbhay Kashyap.
SMART VEHICLE IDENTIFICATION SYSTEM USING OCR in
3rd IEEE International Conference on "Computational
Intelligence and Communication Technology.

11. Khanna, Rishi Anand, “IoT based Smart Parking System”,
2016 International Conference on Internet of Things and
Applications (IoTA) , Pune, Maharashtra.
12. Thanh Nam Pham, Ming-Fong Tsai, Duc Binh Nguyen,
Chyi-Ren Dow, Der-Jiunn Deng, “A Cloud-based Smart-
Parking System based on Internet-of-Things Technologies,
2015 IEEE
13. Rakesh Kumar Lenka, Rabindra K. Barik, Nihal Kumar
Das, Kriti Agarwal, Bebesh Mohanty,SwatiVipsita,“PSPS: An
Iot based predictive smart parking system”, 4th IEEE
conference Uttar Pradesh 2017
14. Rachapol Lookmuang ; Krit Nambut ; Sasiporn
Usanavasin; “ Smart parking usingIoTtechnology”,2018 5th
International Conference on Business and Industrial
research (ICBIR)
15. Basavaraju S R, “Automatic Smart Parking System using
Internet of Things (IoT), International Journal of scientific
and Research Publications, Vol 5, Issue 12, Dec 2015

IRJET- Smart Parking System using Facial Recognition, Optical Character Recognition and Internet of Things (IoT)

More Related Content

What's hot (20)

Similar to IRJET- Smart Parking System using Facial Recognition, Optical Character Recognition and Internet of Things (IoT) (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Smart Parking System using Facial Recognition, Optical Character Recognition and Internet of Things (IoT)