0% found this document useful (0 votes)
34 views5 pages

Parking Space Detection

Successful implementation of Parking Space Detection can effectively reduce a lot of problems related to traffic congestion in urban areas. Wastage of fuel and time in search of a vacant parking space will be significantly reduced and the complete utilization of the available parking areas canmake our cities really smart. Parking slot detection and user notification are the two major sections of a smart parkingsystem.

Uploaded by

SMARTX BRAINS
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
34 views5 pages

Parking Space Detection

Successful implementation of Parking Space Detection can effectively reduce a lot of problems related to traffic congestion in urban areas. Wastage of fuel and time in search of a vacant parking space will be significantly reduced and the complete utilization of the available parking areas canmake our cities really smart. Parking slot detection and user notification are the two major sections of a smart parkingsystem.

Uploaded by

SMARTX BRAINS
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Journal Publication of International Research for Engineering and Management (JOIREM)

Volume: 05 Issue: 09 | Sept-2021

Parking Space Detection


Pratiksha R. Jagtap Adarsh S. Ripote Jayesh D. Pawar Aadesh R. More
Department of Computer Department of Computer Department of Computer Department of Computer
Engineering Engineering Engineering Engineering
Brahma Valley College of Brahma Valley College of Brahma Valley College of Brahma Valley College of
Engineering and Research Engineering and Research Engineering and Research Engineering and Research
Institute, Nashik, MH Institute, Nashik, MH Institute, Nashik, MH Institute, Nashik, MH
[email protected] [email protected] [email protected] [email protected]

---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract for a parking space, travelling at an average of 10 mph and
covers only half a mile in the meantime. The result is frequent
traffic congestion. When the drivers are in search of a parking
Successful implementation of Parking Space Detection space, the possibility for accidents increases as they give less
can effectively reduce a lot of problems related to traffic attention to the road [1]. A sophisticated car parking system
congestion in urban areas. Wastage of fuel and time in can only solve these problems. That is why numerous research
search of a vacant parking space will be significantly works are taking place in this area all around the world. Empty
reduced and the complete utilization of the available parking slot detection is the first phase of any smart parking
system. The second phase is sharing this information to the
parking areas can make our cities really smart. Parking drivers who are in search of parking lots. There are many
slot detection and user notification are the two major methods used for detecting empty parking slots [2].
sections of a smart parking system. The empty parking
space detection was initially done by deploying a number
of sensors in the parking lot. It is highly expensive and RELATED WORK:
complicated to install. But the advancement of image
Smart solutions for car parking in urban areas and the
processing has enabled us to use images of the parking related technologies have been improved tremendously
area to find out empty spaces. In this paper a comparative over the past few decades. Wherever there is a significant
study of the various parking space identification amountof traffic some form of car parking system is also
techniques has been done. Also the image processing needed. Rapid rise of urban population and the increase in
traffic congestion has escalated the demand for the
based system models has been presented as a replacement
researches in the smart parking system domain during the
for sensor based systems. A very efficient and simple past two decades. By taking the advantage of some modern
technique for parking slot identification based on optical tools in information and communication engineering, the
character recognition (OCR) has been introduced in this parkingspace information can be shared effectively in real
paper. The camera installed in the parking lot captures time.
the image of the parking area with specially numbered For example, Parking Guidance and Information (PGI)
parking spaces. The OCR system recognizes the numbers systems [3]-[5] through variable message signs provide
which are not hidden by a vehicle parked over it and this real time parking information to the drivers who are in
search of a parking lot. In order to gather information
information is used for identifying the empty spaces. about occupancy of parking spaces the system uses special
sensorswhich are deployed mainly at the entry gates and
exit gates of the parking lot. These types of systems can
be deployedin the car parking lot of commercial shopping
Keywords: -Parking space detection, Image Processing, malls and trade centers and can be installed easily
Optical Character Recognition
The remaining portion of this paper is organized as follows:
INTRODUCTION: Section III explains the overall system model of a typical smart
parking system. A study of different sensors detecting the
So many researches on traffic congestion analysis reveal empty parking spaces is done in this section. In Section IV, our
that an estimated 70 per cent of all drivers currently on the proposed OCR based system is introduced as an efficient
road are searching for effective parking. This will intensify alternative for existing algorithms. Results and conclusions are
traffic congestion as the vehicles spend more time on the road.
given in the last two sections.
Drivers may also tend to drive at low speed when they are
searching for a parking space. Researches in this area have
found out that vehicles spend an average of 15 minuteslooking SYSTEM MODEL:

© 2021, JOIREM |www.joirem.com| Page 1


Journal Publication of International Research for Engineering and Management (JOIREM)
Volume: 05 Issue: 09 | Sept-2021

A modern Parking Space Detection will have two


fundamental sections such as (A) Empty Parking Spot
Detection section and (B) User notification section. In
the typical system model (shown in Fig.1) we can see that
the parking lot is divided into separate parking spaces [6]. The
B. Parking detection is done at the parking lot and it will
empty parking space is to be identified and the information
find out the occupancy of a vehicle in a parking
about that particular parking area and the available empty space. The presence of a vehicle in the parking space
parking spaces is to be passed on to the driver who is in search is identified with the help of a variety of techniques
of a parking lot. This is done in the user notification stage. The using some type of sensors [6]. Even though the
parking detection system of the parking lot is connected to a techniques and sensors are different the output of
this section will be always the same. It will be a
web server through a gateway. The database stores the parking
binary value whose states indicate the presence and
space information of the entire city. the absence of a vehicle in that particular spot.
Performance of different sensors used for the
This data is given to the driver simply by the help of a mobile detection ofthe presence of a vehicle will depend on
application. The application will guide the driver towardsthe various factors. In the following subsections we will
available parking spaces in the nearby localities examine the role of some sensors for occupancy
detection.

Fig. 1. Schematic Model of a Parking Space Detection

Modern Parking Space Detectioncan be broadly divided into Inductive Proximity sensor:
two based on the technology used for identifying the empty An inductive proximity sensor can detect metal objects
parking spaces in a parking lot; (1) Sensor based methods and approaching it by using the principle of mutual induction.
(2) Image processing based methods. Here in this paper we It can be deployed at the entrance and exit of the parking
make a comparative study about different parking spot lot for identifying the entry and exit of vehicles in the
detection techniques and we also propose an advanced parking parking area. This sensor is usually buried underground at
spot identification technique using OCR algorithm of image the entrance as well as the parking space. Most of the
processing. Inductive Proximity Sensors usea coil and an oscillator for
producing an electromagnetic field in the close
surroundings. The vehicle which is a metal body can cause
a change in the magnetic field which will in fact cause a
damping of oscillation amplitude. The rise or fall of such
oscillation beyond a given threshold value changes the
output of the sensor. This sensor will only detect the
A. Sensor Based Methods presence of metal bodies such as vehicles and will be

© 2021, JOIREM |www.joirem.com| Page 2


Journal Publication of International Research for Engineering and Management (JOIREM)
Volume: 05 Issue: 09 | Sept-2021

inactive if some living or nonmetal objects enter its Image processing based parking space detection proposed
proximity. The main demerit is the complexity in the by Yusnita and Norbaya and so many other similar papers
installation and maintenance of the sensors in the parking introduce algorithms using imagesegmentation and detection
lot. techniques. Here a round white image will be drawn on each
parking slot manually. This will enable the system to
automatically identify the location of every parking slot in the
Active Ultrasonic Sensor: image. The absence of the white circle in the image will be
considered as the occupancy of a vehicle in the parking space.
a) A special sensor that can be used for detecting the
presence of a vehicle is an ultrasonic sensor. These sensors
will generate ultrasonic pulse waves which are used in
them for measuring the distance between the object and
the sensor. The sensor transmitter sends ultrasonic pulses
and the reflected pulses are received by it for measuring
the time delay. By using this time delay between the
emission and reception, the distance betweenthe sensor
and the target will be calculated. The ultrasonic sensor can
accurately detect the presence of objects in the parking
spaces but it has no provision to distinguish between the
occupancy of a vehicle and some other objects. That is,
even the presence of an animal or a person will be treated Fig. 3. Schematic of Image Processing Based Parking Space
as a vehicle occupying the space. Also, the system is very Detection.
costly as we need sensors for each and every parking spot.
The installation and maintenance of the system is
complex.
The first module is system initialization which will
automatically set up the boundaries of every parking space in
RFID Sensor: the entire parking area. A camera is used for capturing the
Radio Frequency Identification (RFID) ) isused for automatic image of the parking area and then it is stored in the memory.
identification of objects with the help of radio waves. RFID It is taking place in the image acquisition section. The
uses a tag or a chip which will be installed in the vehicle and acquisition unit is connected to an image processing unit that
will send a code which will be received by an RFID reader runs a MATLAB program. The third module is image
present in the parking area and the data can be used for segmentation, which uses thresholding technique. Then the
accurately identifying the arrival of a vehicle. When the image is enhanced by using morphological functions which
vehicle enters the parking area, the data stored in the RFID will remove pixels that does not belong to the selected object.
chip will be received by the RFID reader. Only vehicles with Image detection is the last module, which identifies the round
correct RFID tag will be identified. Vehicles without the brown image on each and every parking space. Based on the
RFID tag will be blocked from entering the parking area. output of image detection section the empty parking lots can
Although this a very simple and effective technique, the be identified. Fixing of the camera at a proper location is very
installation of RFID tags in each andevery vehicle is a highly critical in this method. Any variation in the orientation of the
complex process. camera will cause errors in the output.

LIDAR Sensor: PROPOSED SYSTEM:

a) LIDAR (Light Detection and Ranging) System Initialization


is similar to that of ultrasonic sensors but uses light pulses
instead of RF pulses. Although this sensor is more accurate A one-time manual drawing of slot numbers is done in
even at longer distances it is highly expensive. the parking area as shown in figure 4. In order to identify
the location of each parking space a unique number is
assigned to each and every parking slot. A camera is set up
Sensor based methods have many disadvantages such as high in a position to capture the entire slot numbers in a single
installation cost per parking area, complexity of the system frame. Unlike the earlier systems, there is no need to give
causes confusion for customers and there may be a chance of extra care for fixing the camera. The only one thing here
breakdown which requires a maintenance contract with the we need to ensure is that the captured image should contain
supplier. all the slot numbers clearly (as shown in Fig.4).

Image Processing Based Methods:

© 2021, JOIREM |www.joirem.com| Page 3


Journal Publication of International Research for Engineering and Management (JOIREM)
Volume: 05 Issue: 09 | Sept-2021

slots numbers in a single frame. There is a five second delay


to take images by camera. Then the image is stored for
further processing.
b) Processing the image: The captured image is processed
for improving the quality of the image. Basically it involves
RGB to gray scale conversion and contrast enhancement.
The image processing done in this stage will also eliminate
unwanted noises present in the captured image. This will
create an enhanced image for further processing.
c) Image Segmentation and Recognition: The image
segmentation is the key technique used in optical character
recognition based system. This section focuses on
identification of the slot numbers. A MATLAB based OCR
Fig. 4. Image of the empty parking lot with individual algorithm is used for the recognition of the numbers in the
numbers for each slot. image. These numbers are stored as a vector.
d) Identifying the empty slots: If a vehicle is parked in a slot
then its slot number will be hidden. So the slot number will
not be present in the output vector. The missing number is
identified as an occupied slot. The available slot numbers in
the output vector represents the empty slots and it will be
transferred to the user notification stage. These numbers can
be displayed as vacant spaces in the parking area.

RESULTS:

The image of the parking area is processed to obtain the


gray scale image and then it is converted into the
corresponding binary image. After proper filtering, the image
Fig. 5. Image of the parking lot with cars parked. Slot numbers will show only the slot numbers as shown in Fig. 7.
are only visible in the empty slots.

When cars are parked in the lots the corresponding slot


number will be hidden in the image. Now the only one
thingwe need to do is to identify the available slot numbers.
This can be done by using Optical Character Recognition
(OCR) techniques of digital image processing. A typical
OCR based number recognition model is shown in Fig. 6.

Fig.7. Binary image of the parking area contains only empty


parking slot numbers

This image is processed by using OCR function of the


MATLAB which will result the characters present in the
image. The characters are converted into integer numbers
for further processing and passed on to the user notification
section. The accuracy of the empty parking slot detection is
very high when we use this method. It is seen that by using
proper filtering before and after banalization the number
System model detection accuracy can be further increased.

a) Capturing the image of the parking area: The camera


A GUI window has been created in MATLAB for
is used to captures the image of the parking area. It is placed displaying the status of each and every parking slot in the
ata particular angle to monitor the entire parking area and the

© 2021, JOIREM |www.joirem.com| Page 4


Journal Publication of International Research for Engineering and Management (JOIREM)
Volume: 05 Issue: 09 | Sept-2021

parking area. A typical output GUI window is shown in


Fig.8.When we give the image of the parking lot as the input
of the MATLAB code, the GUI window will display the
status of all the slots in the parking area. It will be either
‘occupied’ or ‘empty’.

CONCLUSION:

The first section of the Parking Space Detection is the


parking slot detection. This can be done using sensors
placed at different locations of the parking lot. But this is a
very costly system with a lot of drawbacks. The image
processing based empty parking lot identification has made
the system simple as well as cheap. Here we need only to
use images captured by the surveillance camera in the
parking lot for empty slot detection. OCR based solution
will be an integral part of future smart parking systems.
Using suitable image processing algorithms, the accuracy
of detection can be improved.

REFERENCES:
[1] T. Lin, H. Rivano and F. Le Mouël, "A Survey of
Smart Parking Solutions," in IEEE Transactions on
Intelligent Transportation Systems, vol. 18, no. 12,
pp. 3229-3253, Dec. 2017.
[2] N. Bibi, M. N. Majid, H. Dawood and P. Guo,
"Automatic Parking Space Detection System," 2017
2nd International Conference on Multimedia and
Image Processing (ICMIP), Wuhan, 2017, pp. 11-15.
[3] Y. Ji, W. Guo, P. Blythe, D. Tang, and W. Wang,
“Understanding drivers’ perspective on parking
guidance information,” IET Intell. Transport Syst.,
vol. 8, no. 4, pp. 398–406, June 2014.
[4] H. Guan, L. Liu, and M. Liao, “Approach for planning
of parking guidance and information system,” J.
Highway Transport. Res. Develop., vol. 1, pp. 034,
2003.
[5] F. Caicedo, “Real-time parking information
management to reduce search time, vehicle
displacement and emissions,” Transport. Res. D, vol.
15, no. 4, pp. 228–234, 2010.
[6] R. Grodi, D. B. Rawat and F. Rios-Gutierrez, "Smart
parking: Parking occupancy monitoring and
visualization system for smart cities," SoutheastCon
2016, Norfolk, VA, 2016, pp. 1-5.

© 2021, JOIREM |www.joirem.com| Page 5

You might also like