www.ijcrt.

org © 2021 IJCRT | Volume 9, Issue 11 November 2021 | ISSN: 2320-2882

AUTOMATIC NUMBER PLATE RECOGNITION

SYSTEM FOR VEHICLE IDENTIFICATION
USING MACHINE LEARNING

Ms. Sayali Lokhande1, Ms.Pragati Sahane2 , Ms. Sonali Katare3, Ms. Dnyaneshawari Tawhare4 , Mrs.Sweta Khandekar

for character recognition tool to translate the pixels into

numerically readable character. ANPR can be used in
Abstract—Automatic Number Plate Recognition many areas from speed enforcement and tool collection
(ANPR) is an image processing technology which uses to management of parking lots, etc . It can also be used
number (license) plate to identify the vehicle. The to detect and prevent a wide range of criminal activities
objective is to design an efficient automatic authorized and for security control of a highly restricted areas like
vehicle dentification system by using the vehicle military zones or area around top government offices.
number plate. The system is implemented on the The system is computationally inexpensive compare to
entrance for security control of a highly restricted area the other ANPR systems . Besides the robustness, the
like military zones or area around top government earlier methods use either feature based approached using
offices e.g. Parliament, Supreme Court etc. The edge detection or Hough transform which are
developed system first detects the vehicle and then computationally expensive or use artificial neural
captures the vehicle image Vehicle number plate region network which requires large training data . The
is extracted using the image segmentation in an image. presented ANPR system is aimed to be light weighted
Optical characte r recognition technique is used for the so that it can be run real time and recognizes
character recognition. The resulting data is then used to Sindhstandard number plate under normal conditions.
compare with the records on a database so as to come The ANPR system works in three steps, the first step is
up with the specific information like the vehicle’s the detection and capturing a vehicle image, the second
owner, place of registration, address, etc. The system is steps is the detection and extraction of number plate in
implemented and simulated in Matlab, and it an image. The third section use imagesegmentation
performance is tested on real image. We proposed the technique to get individual character and optical
automatic vehicle identification system using vehicle character recognition (OCR) to recognize the individual
license plate is presented. The system use series of character with the help of database stored for each and
image processing techniques for identifying the every alphanumeric character. The rest of the paper is
vehicle from the database stored in the PC. The system organized as follows: will present the software and
is implemented in Matlab and it performance is tested hardware models of the developed ANPR system. will
on real images. The simulation results shows that the present the simulation results obtained using the
system robustly detect and recognize the vehicle using developed ANPR system. Section 4 discus ses the results
license plate against different lightening conditions and briefly and finally will end the paper with conclusion
can be implemented on the entrance of a highly and future works.
restricted areas.
II.Related work(Literature Survey)
K.K. Kim et. al (2000) built a license plate
Introduction:The Automatic Number Plate recognition system by following a learning protocol.
Recognition (ANPR) was invented in 1976 at the The camera captures an image inside the car
Police Scientific Development Branch in the UK. detection module. Then, the picture of the candidate
However, it gained much interest during the last decade region is provided as output. The two TDNNs were
along with the improvement of digital camera and the
taken as the horizontal andvertical filters to find the
increase in computational capacity. It is simply the
license plate. The segmentation rate was 97.5 per
ability to automatically extract and recognition a
cent, and a recognition rate of 97.2 per cent for the
vehicle number plate’s characters from an image. In
proposed system .
essence it consists of a camera or frame grabber that
 Chin-Chuan Han et. al (2007) suggested a
has the capability to grab an image, find the location of
the number in the image and then extract the characters device that not only tracks several targets but also
obtains highquality images on plate numbers. A

IJCRTI020012 International Journal of Creative Research Thoughts (IJCRT) www.ijcrt.org 47

www.ijcrt.org © 2021 IJCRT | Volume 9, Issue 11 November 2021 | ISSN: 2320-2882

computer with a tuned dual-camera system has been the number plate region and then moves it to the
built here by the author; a stationary camera and a second YOLO model for segmentation and platform
pan-tilt-zoom camera are designed to monitor image recognition. As a result, the model was checked
moving conveyance in an open field. The license with 200 images and correctly recognized the license
plate for recognition has been sequentially identified plate number for 199 images, i.e. 99.5 percent accuracy
by CNN classifier. As 64 vehicles entered this rate Shally Gupta, Department of Information
region illegally, data was composed manually from Technology, College of Technology, Govind Ballabh
the science pictures and 59 IDs were Pant University of Agriculture and Technology,
accurately India, E-mail: [email protected]
detected using this Received date: July 01,
tool 2020; Accepted date: July 20, 2020;
 Madhusree Mondal et. al (2017) Published date: July 28, 2020 This paper describes a
developed an ANPR framework focused on the study of vehicle number plate identification in traffic
learning capabilities of convolutional neural surveillance. ANPR is very helpful and reliable for
networks. The self-synthesized function of CNN efficient traffic
was aAused here, as it distinguishes the vehicle monitoring. Device with powerful image processing
states from the number plate. technique can easily detect interested vehicles from
The system was organized in this work in an echelon various angles and view owner information as
network of feature detectors that conducted consecutive output. ANPR systems play an important role in the
processing of visual data pertaining to the dominant growth of the smart transport network. Recognition
visual processing experience of the visual cortex, which may use the image processing technique combined
influenced the computational model of the CNN. The with neural networks to identify the number plates
findings of this research were observed with fewer where the angled or side view images, moving
training samples and turned out to be as 90 per cent distance images,numbering scheme, and number
higher precision rate plate type (background) can be further improved.
Andrew S. Agbemenu et. al (2018) Detection of objects and neural networks is useful
proposed an ANPR method based on the characteristics for detecting side views or tilted images and moving
and variations of the plates therein. The author has images from distance.
proposed in this work an algorithm that is enhanced to
perform with Ghanaian license plate for conveyance. III. Flow Diagram
The designed model used two candidate detection
algorithms as the detection of edges and the algorithms
matching the template. The device then implemented
the character segmentation technique particularly with
square plates to prevent noise effects, arrangement of
characters and skewing. At the final point, character
recognition was rendered with the use of tesseract OCR
engine. Feature detection was slightly low but had a
good success rate, with an average speed of 0.185s
detecting 454 plates with 90.8 per cent accuracy. The
optical character recognition provided an average of
0.031s for the procedure and successfully identified
approximately 60 per cent of the detected plates .
Nazmus Saif et. al (2019) have proposed a system
to detect and recognize the Bangla license plate from
the vehicle picture by using the convolutional neural
networks. In this work, main focus to choose IV. Proposed System
convolution neural network in the designed system is
To recognize number plate first of all add templates
preferred because of its configuration for the end-to-
from A-z and 0-9 and add them into mat file. After
end pipeline. CNN clearly outperformeconventional
that read the image and change over that image into
image processing algorithms for their case, and
gray scale. Presently the following stride is to find
compared generalized CNN models better in different
out threshold estimation of the image. In the wake of
scenarios. The detection research was done using
finding T-esteem change over that image into binary
YOLOv3 which consists of 53 convolutional model
layers. The second stage after identification is image
V. Conclusion
segmentation and recognition of the characters it is.
a study of vehicle number plate identification in
During this step, the device whips out
traffic surveillance. ANPR is very helpful and
reliable for efficient traffic monitoring. Device with
IJCRTI020012 International Journal of Creative Research Thoughts (IJCRT) www.ijcrt.org 48
www.ijcrt.org © 2021 IJCRT | Volume 9, Issue 11 November 2021 | ISSN: 2320-2882

powerful image processing technique can easily Angles Using Mask RCNN. In 2019 International
detect interestedvehicles from various angles and Conference on Advanced Mechatronic Systems
view owner information as output. ANPR systems (ICAMechS) (pp.
play an important role in the growth of the smart 229-234). IEEE.
transport network. Recognition may use the image
processing technique combined with
neural networks to identify the number plates where
the angled or side view images, moving distance
images,numbering scheme, and number plate type
(background) can be further improved. Detection
of objects and neural networks is useful for detecting
side views or tilted images and moving images from
distance. is to use high-resolution cameras with an
increased . number of frames for good accuracies and
improved accuracy for recognition in future.

VI. References
1. Kim, K.K., Kim, K.I., Kim, J.B. and Kim,
H.J., 2000, December. Learning-based
approach for license plate recognition. In Neural
Networks for Signal Processing X. Proceedings of
the 2000 IEEE Signal Processing Society
Workshop (Cat. No. 00TH8501) (Vol. 2, pp. 614-
623). IEEE.
2. Han, C.C., Hsieh, C.T., Chen, Y.N., Ho,
G.F., Fan, K.C. and Tsai, C.L., 2007, October.
License plate detection and recognition using a
dual-camera module in a large space. In 2007 41st
Annual IEEE International Carnahan Conference
on Security Technology (pp. 307-312). IEEE.
3. Mondal, M., Mondal, P., Saha, N. and
Chattopadhyay, P., 2017, December. Automatic
number plate recognition using CNN based self-
synthesized feature learning. In 2017 IEEE Calcutta
Conference (CALCON) (pp. 378-381). IEEE.
4. Agbemenu, A.S., Yankey, J. and Addo, E.O.,
2018. An automatic number plate recognition
system using opencv and tesseract ocr engine.
International Journal of Computer Applications,
180, pp.1-5.
5. Laroca, R., Severo, E., Zanlorensi, L.A.,
Oliveira,
L.S., Gonçalves, G.R., Schwartz, W.R. and Menotti,
D., 2018, July. A robust real-time automatic license
plate recognition based on the YOLO detector. In
2018 International Joint Conference on Neural
Networks (IJCNN) (pp. 1-10). IEEE.
6. . Dhar, P., Guha, S., Biswas, T. and Abedin,
M.Z., 2018, February. A system design for license
plate recognition by using edge detection and
convolution neural network. In 2018 International
Conference on Computer, Communication,
Chemical, Material and Electronic Engineering
(IC4ME2) (pp. 1-4).
IEEE.
7. Lin, C.H. and Li, Y., 2019, August. A
License Plate Recognition System for Severe Tilt

IJCRTI020012 International Journal of Creative Research Thoughts (IJCRT) www.ijcrt.org 49