Automatic Number Plate Recognition System for

Entry and Exit Management

A.Balaji
M r.Vijaya Vardan Reddy SP B.Sathyasri
Department of Electronics and
Department of Electronics and Department of Electronics and Communication Engineering
Communication Engineering Communication Engineering Rajalakshmi Institute of
R.M.K.Engineering College Vel Tech Rangarajan Technology
Chennai, India Dr.Sagunthala R&D Institute Chennai, India
[email protected] of Science and Technology [email protected]
[email protected]
2021 6th International Conference on Communication and Electronics Systems (ICCES) | 978-1-6654-3587-1/21/$31.00 ©2021 IEEE | DOI: 10.1109/ICCES51350.2021.9489018

Rahul Krishnan Y.Deepika

S.Vanaja Department of Electronics and Testing Engineer
Department of Electronics and Communication Engineering TATA CONSULTANCY SERVICES.
Communication Engineering Rajalakshmi Institute of CHENNAI
Easwari Engineering College Technology [email protected]
Chennai, India Chennai, India
[email protected] [email protected]

Abstract–The main objective of the system is to monitor the

vehicles that are entering and going out of the organization. All
vehicles have their own unique license plate number, so the
abstraction of plate number plays a major role in this system.
The vehicle number plate is taken by the tool like camera which Nowadays this ANPR play an important role in
is placed at the entrance. The captured image is then processed traffic safety enforcement, automatic toll car park system [2],
by the Automatic Number Plate Recognition (ANPR), an image text collection [3], Automatic vehicle parking system [4],
processing algorithm. ANPR method is an efficient way of detection of car speed [5], restricted areas security control
recognizing the vehicle number plate and strengthen the security process [6], unattended parking zones [7], electronic toll
system. A database is created with the vehicle number which
collection [8],traffic law enforcement [9] etc. The most
belongs to that particular organization. The recognized number important processing phases involved to achieve the proposed
plate is then compared with the database and checks whether the
project is listed below. 1. Image acquirement 2. Recognize the
vehicle belongs to the organization if not then it is an unknown
number plate 3. Detect the edge from the acquired signal 4.
vehicle. Then the entry and exit time of the organization vehicles
Segment the extracted character 5. Recognize the character
are recorded. In case of centralized receiver all the entrance
records are kept in storage and any unfamiliar vehicle entering
and corresponding with database [10].
into the organization is noticed and central controller monitor
In world-wide, all vehicles must have their license
the details.
number is to be displayed on the license plate which is
Keywords-– Character Recognition, Automatic Number Plate attached at the front and back side of the vehicle. But the issue
Recognition, Vehicle Number Plate, Raspberry Pi3, processor, is some vehicle whose license number is not visible properly,
VNC Viewer, Camera, Character S egmentation. not correctly mounted.
In majority of the organizations like colleges, schools
I. INT RODUCT ION and the IT sector the major issue is security determination.
In recent years, license plate recognition (LPR) or Majority of the administration will maintain the exit and entry
ANPR has played an important role in vehicle surveillance of the vehicle. In majority of the organization the exit and
system [1] is illustrated in the Fig. 1. entry record is maintained manually by the security person of
the administration. The security person will add the entry and
exit time of each vehicle. Hence in order to overcome these
difficulties ANPR system is proposed. This technique will
deliver more clarity and efficiency for preserving these
records.
In present situation license plate recognition (LPR)
scheme plays a vital in the field of vehicle parking system,
Electronic toll collection, security purpose, speed of the car
etc.
For the identification of the vehicle individually we
can use an RFID tag and IoT based application [11-18]. But
this reader has certain disadvantages such as, operates with in
minimum distance, implementation is difficult & time
consuming, less reliable, scanning issues etc.

Fig. 1 Example of ANPR System

Authorized licensed use limited to: UNIVERSITI TEKNOLOGI MARA. Downloaded on April 17,2024 at 09:58:09 UTC from IEEE Xplore. Restrictions apply.
 To overcome the above limits, the proposed system number plate using preprocessing methods. The number plate
have high pixel quality camera is used to gather the image of captured image is converted into text.
the vehicle which is entering and exiting and then through
image processing technique recognize number for each and
every vehicle into the organization. The high pixel camera is
attached to the high speed processor (raspberry pi 3) in order
to perform the image processing technique.
Furthermore, the proposed system specifically
concentrates on the educational institutions in which the
system will be placed at the entrance of the organization. The
system extracts the number plate from the image after doing
the image processing steps. This data is stored in a database to
keep out the daily records. This information provides a better
clarification for the organization in maintaining their records
about the entry and exit time of vehicles entering into the
organization. The information will be finally stored in the
local server and these records can be accessed from anywhere
such that a uniform resource locator (URL) is provided with
recorded information.

II. MET HODOLOGIES

In this section the methodology of the proposed ANPR is

explained in detailed as shown in Fig. 2. The raspberry pi 3 Fig. 3 Entry and exit system
acts as a central processor. A webcam is connected to a
raspberry in order to capture the image at the entrance of the (i) ENTRY System
organization. Raspberry pi 3 is connected to the network to
transfer the extracted information to the database server. Initially, the execution starts by capturing the vehicle by
Initially the image is captured when the licensed plate is camera and Vehicle Number Plate is extracted from the
detected. This captured image is fed into the processor for captured image. From the Vehicle Number Plate image,
processing the image. Raspberry pi 3 is operated using Virtual number image alone extracted and it is converted into text.
network computing (VNC) viewer. It is a variety of remote- Whenever the captured number plate matched with the
control software that control extra computer through a network database number plates images then it comes under the
connection. The raspberry is coded then the image is done organizational vehicles. Then the matched number plate is
with several processing steps. Then finally the information is stored in the database along with the entry time of the vehicle
stored in a database with their respective time of vehicle entry at the entry gate. On condition that, number plate is not go
to the organization. This information can be accessed by with the number plates stored in the database, then that
everyone in that organization by providing a URL link to number plate will be categorized under Outsider’s Vehicle.
access the webpage. Hence entry time alone is noted. Finally the entry system
execution stopped.

(ii) EXIT System

The same process is pursued during the initial stages like

Acquisition, preprocessing and converting image to text.
When this process is completed, the exit time (time of
capturing the image of the vehicle at the exit gate) is saved.
Then this exit execution stopped.

III. RESULT S AND DISCUSSIONS

The proposed methodology is to acquire the license
plate number of the vehicle entering the organization and store
entry time and exit time in the database without any human
intervention. The following steps to be followed for ANPR
system. a) Image acquisition b) Image preprocessing c)
Localization of number plate d) Character segmentation e)
Fig. 2 ANPR basic block diagram
Character recognition f) Storing in and out time in database.
The workflow of the proposed Entry and Exit system is shown
A. Image Acquisition
in the Fig. 3. Image Acquisition is done when the vehicle is
entered in to the organization and enhances the Vehicle
Logitech c525 camera is used to capture the
image of the vehicle at the entrance as shown in the Fig. 4.

Authorized licensed use limited to: UNIVERSITI TEKNOLOGI MARA. Downloaded on April 17,2024 at 09:58:09 UTC from IEEE Xplore. Restrictions apply.
The webcam is connected to the processor of the system. The shape, this region of interest is obtained by comparing with the
ultrasonic sensor gives a trigger pulse to the camera whenever different dimensionality of the edge. Contour detection
it detects the vehicle and the camera captures the image and method is followed, which extracts all the closed shapes or
sends it to the processor. boundaries from the image. In the proposed work license plate
is in rectangular in shape, extract the rectangle shape from the
license plate region using filtering process. Then,
morphological operations are performed to remove the
unwanted information from the image as shown in Fig. 6.
Edge detection process is applied in both vertical and
horizontal regions of the number
plate.

Fig. 4 Acquired image

B. Image Preprocessing

Image preprocessing is followed to eradicate

the added noise and distortion present in the captured image.
Here we use Gaussian filter and adaptive thresholding to
preprocess the image. The Gaussian filter is faster than other
filters and it remove the noise in the image but the resulting
image is in blurred state. Adaptive thresholding process
converts the color image to the gray scale image as shown in Fig. 6 Localization of number pate
the Fig. 5.
D. Character Segmentation

The characters are truncated and stored in which the

process is continued until no characters remain in the region of
plate as shown in Fig. 7. Each and every character is stored in
the separate variable in order to recognize it separately. The
size of the defined boxes varies and which depends on the size
and shape of the character.

Fig. 5 Processed gray scale image Fig. 7 Segmentation of number pate

In input image is made of pixel in which the intensity of each E. Character Recognition
pixel image ranges from 0 to 255. All the pixel values are
added and then take average value. This average value In optical character recognition (OCR) system,
considered as threshold value. The resultant threshold value is Python-tesseract is an important tool used in python. It will
compared with each and every pixel present in the image. If read and recognize the text information present in the images.
the value of pixel present in the image is greater than the If it is used as a script then it will print the recognized text
threshold value than binary 1 is assigned to the particular pixel instead of writing it to a fileThe character is using the
otherwise assign binary 0. Binary 1 represents white color technology of Optical Character Recognition (OCR) which
whereas binary 0 represents black color. In this way, color extracts the characters present in the license plate. Once the
image is converted into gray scale image. license plate is recognized the obtained region is created as a
new image. OCR is applied to the new image and it extracts
C. Number Plate Localization all the characters. The characters may be the numbers and
alphabets present in the number plate. Then the obtained
The preprocessed image is applied to the edge image is converted to text and it is stored in a string variable.
operator in order to extract the license plate region from the Initially the training images having different characters are
background. Generally the license plate is in rectangular processed so that the system will understand the different

Authorized licensed use limited to: UNIVERSITI TEKNOLOGI MARA. Downloaded on April 17,2024 at 09:58:09 UTC from IEEE Xplore. Restrictions apply.
features and characters. Finally the grey image is converted Further, the number plate of the vehicles entry and exit time,
into RGB format with the recognized number plate highlighted altogether can be viewed perfectly by creating the Basic Front-
with a red border as shown in Fig. 8 and the resultant string End Web application along with the Back-End consisting of a
variable is stored in the database. Using PyTesseract the database for the storage of organizational vehicles Number
characters present in the boundary are converted to text. The plate as shown in Fig. 10. The data can be categorized by date,
converted text is easily stored in the database for farther month and year as per the user convenience. Web page can be
utilization. The converted text can be easily viewed in the the accessed from anywhere, so dynamically it can be accessed.
excel format. The number plate along with entry and exit time The vehicle management will become easy for the
is viewed clearly by Basic Front-End Web application. organization by using this proposed application. This System
will build up the security purpose of the organization.

Fig. 8 Character Recognition

F. Storage In Database

For storing the resultant in a database MySQL Fig. 10 Webpage of ANPR process
package is used. MySQL connector package links the python
directly with the database. Also the package ‘date and time’
will display the current time of capturing. The Text can be IV. CONCLUSION
easily viewed in the Excel as shown in Fig. 9. Hence all the
reports of the organization vehicle and outsider vehicle are The proposed work focus on the recognition technology where
the image of the vehicle plate is acquired using digital cameras
stored in the database. The outsider vehicles are identified by
comparing with the initially created database. In this database and the resulting image is undergone many preprocessing
all the license number of the vehicles belonging to the steps in order to get information about the number plate. This
proposed work deals with intellectual digital image processing
particular organization are collected and stored separately. If
the vehicle is matched with the database then the vehicle method to be familiar with the English digits and numerals. A
number is stored in the field of authorized vehicle with its back end image is seized and undergone various image
processing steps using numerous algorithms. In future, analyze
entry date and time, else the vehicle number is categorized as
unauthorized vehicle and data and time is stored in the the characteristics of ANPS in order to obtain the better
separate field. performance. In future this ANPS provides a mass
surveillance and man less security system for the organization.

REFERENCES

[1] https://www.pyimagesearch.com/2020/09/21/opencv -automatic-

license-number-plate-recognition-anpr-with-python/
[2] C. Anton Satria Prabuwono and Ariff Idris, "A Study of Car Park
Control System Using Optical Character Recognition," in
International Conference on Computer and Electrical Engineering,
(2008): 866-870.

[3] You-Shyang Chen and Ching-Hsue Cheng, "A Delphibased rough

sets fusion model for extracting payment rules of vehicle license tax
in the government sector," Expert Syst ems with Applications, vol.
37, no. 3, (2010): pp. 2161- 2174.

[4] A Albiol, L Sanchis, and J.M Mossi, "Detection of Parked Vehicles

Using Spatiotemporal Maps," IEEE T ransactions on Intelligent
T ransportation Systems, vol. 12, no. 4, (2011): 1277-1291.

[5] Chaithra, B., Karthik, K.V., Ramkishore, D. and Sandeep, R.,

Fig. 9 Database storage September. Monitoring T raffic Signal Violations using ANPR and
GSM. In 2017 International Conference on Current T rends in

Authorized licensed use limited to: UNIVERSITI TEKNOLOGI MARA. Downloaded on April 17,2024 at 09:58:09 UTC from IEEE Xplore. Restrictions apply.
 Computer, Electrical, Electronics and Communication (CT CEEC),
IEEE, (2017): (pp. 341-346,.

[6] Kulkarni, P., Khatri, A., Banga, P. and Shah, K., “ Automatic number
plate recognition (anpr) system for indian conditions”, In 2009 19th
International Conference Radioelektronika, IEEE, (2009): 111-114.

[7] Singh, A. and Vaidya, S.P., “ Automated parking management system

for identifying vehicle number plate”, Indonesian Journal of
Electrical Engineering and Computer Science, vol. 13, no. 1, (2019):
77-84.

[8] Ahmed, S., T an, T .M., Mondol, A.M., Alam, Z., Nawal, N. and
Uddin, J., “ Automated T oll Collection System Based on RFID
Sensor.” In 2019 International Carnahan Conference on Security
T echnology (ICCST ), IEEE, (2019): 1-3.

[9] Setoshita, S., Andrei, C., Phillips, S., Salem, H.H., Seidowsky, R.,
Yao, M., Chin, K.K., Chen, K. and Ikeda, Y., “Smart Use of Roads”,
IT F Research reports (Doctoral dissertation, IFST T AR), 2019.

[10] Kaur, S. and Kaur, S., “ An efficient approach for number plate
extraction from vehicles image under image processing.”
International Journal of Computer Science and Information
T echnologies, vol. 5, no. 3, (2014): .2954-2959.

[11] Bhardwaj, D. and Mahajan, S., “ Review paper on automated number

plate recognition techniques.” Int. J. Emerg. Res. Manag. T echnol,
no. 4, (2015): 319-324.

[12] Durani, A.M., Ali, M., Ahmad, R., Irfan, S. and ur Rehman, H.,
“ Identification and Verification of Vehicle using RFID T echnique.”
VAWKUM T ransactions on Computer Sciences, vol. 4, no. 1,
(2016): 36-43.

[13] Sen, E.J., Dixon, K.D.M., Anto, A., Anumary, M.V., Mieheal, D.,
Jose, F. and Jinesh, K.J., “ Advanced license plate recognition system
for car parking.” In 2014 International Conference on Embedded
Systems (ICES), IEEE, (2014): 162-165).
[14] Krishnan, R., Babu, R.G., Lalitha, K., Vanaja, S. and Devnesh, K.N.,
“ Autonomous Underground Water Detection Robot.” In IOP
Conference Series: Mat erials Science and Engineering, vol. 1055, no.
1, p. 012002). IOP Publishing, 2021.
[15] Balaji, A., Sathyasri, B., Vanaja, S., Manasa, M.N., Malavega, M. and
Maheswari, S., “ Smart Kitchen Wardrobe System Based on IoT.” In
2020 International Conference on Smart Electronics and
Communication (ICOSEC), IEEE, (2020): 865-871).
[16] Landaluce, H., Arjona, L., Perallos, A., Falcone, F., Angulo, I., &
Muralter, F. “ A review of iot sensing applications and challenges using
RFID and wireless sensor networks.” Sensors, vol. 20, no. 9, (2020):
2495.
[17] Samuel, A. J., & Sebastian, S. “ An algorithm for IoT based vehicle
verification system using RFID.” International Journal of Electrical
and Computer Engineering, vol.9, no. 5, (2019): 3751.
[18] Akshay, P. M., Murugesh, K., & Patra, Y. “IoT based Automated Paid
Parking Using Electromagnetic RFID T ag.” In 2019 International
Conference on Intelligent Computing and Control Systems (ICCS),
IEEE, (2019): 1451-1452.

Authorized licensed use limited to: UNIVERSITI TEKNOLOGI MARA. Downloaded on April 17,2024 at 09:58:09 UTC from IEEE Xplore. Restrictions apply.