AUTOMATIC

NUMBER PLATE
RECOGNITION
SYSTEM
UNITED COLLEGE OF
ENGINEERING
AND RESEARCH
UNDER THE SUPERVISION
OF MS. PALLAVI SHUKLA

SUBMITTED BY
DEV SIKKA 2000100100058
DEVANSH TRIPATHI 2000100100059
PRATHAM PANDEY 2003420100061
SAKSHAM SRIVASTAVA 2100100109013
CONTENT
01 INTRODUCTION

02 LITERATURE SURVEY

03 FLOW CHART

04 REAL-TIME OBJECT DETECTION

05 IMAGE PROCESSING

06 OPTICAL CHARACTER
RECOGNITION (OCR)
07 TEMPLATE MATCHING
 CONTENT
08 DATA FLOW DIAGRAM (DFD)

09 ENTITY RELATIONSHIP
DIAGRAM
10 PROGRESS OF THE PROJECT

11 CONCLUSION

12 BIBLIOGRAPHY
INTRODUCTI
ON
• Auto matic Number Plate Reco g nitio n
(ANPR) is a mass surv eill an ce sy stem th at
captu res th e imag e o f v ehicles an d
reco gn izes th eir license n umber

• The ANPR is the imag e p ro cessing

techn iqu e to ex tract th e imag e o f th e
license pl ate.

• To iden tify the veh icl es u sin g their nu mb er

p lates we will req u ire Optical Character
Recog n ition (OCR).
 LITERATURE
SURVEY
Here’s an overview of the research that had been done on ANPR

In 2019,Cheng_Hung_Lin proposed a In 2018,AndrewS.Agbemenu presented an Ms.Sushama,H.Bailmare, and

three stage licence plate identification
method based on Mask_RCNN, which
is employed for multitudinous oblique
pictures & colourful firing angles. In
the automobile discovery stage, the
author employed YOLO-v6 for the
associated car.
ANPR system based on the plates'
characteristics and variances. In this paper,
the author proposes an algorithm that is
upgraded to function with Ghanaian
automobile licence plates.
Prof.A.B.Gadicha demonstrated an
approach based on simple and
effective morphological procedures,
as well as the Sobel edge detection
system.
FLOW
CHART
1. Real-time object detection
2. Image Processing
3. Optical Character Recognition (OCR)
4. Template Matching
REAL-TIME
OBJECT
DETECTION
Most important object detection
algorithms-
• YOLO (2016)
• SSD (2016)
• RetinaNet (2017)
• YOLOv3 (2018)
• YOLOv4 (2020)
• YOLOR (2021)
• YOLOv7 (2022)
• YOLOv8 (2023)
IMAGE
PROCESSING
• Op en CV(Open So urce Com pu ter
Visio n) is o ne of the m o st widely u sed
libraries fo r im age-p ro cessin g tasks in
ANPR sy stem s.

• Op en CV is a free cro ss-p latform

com pu ter v ision lib rary for real-tim e
im age p rocessing .
CONTINUE..
Gray Scale Conversion
T h e colo r in form atio n is no ise in im ag e
pro cessin g as it mak es it difficult for the
so ftware to pre-pro cess the im age. On the
other h an d, gray scale has less size and is
accep table in mo st of the alg orithm s.

Binarization
T h e Binarization Method con verts the
im age into an im age in wh ich the p ixel
can o nly hav e 2 v alues – 0 or 1 , i.e., a
black and white imag e.
OPTICAL CHARACTER
RECOGNITION (OCR)

OCR refers to a software techn olo gy th at

electro nically identifies text in sid e an
im age file and con verts it into a m achin e-
readable text form to be u sed fo r data
pro cessin g.
TEMPLATE
MATCHING
Character Segmentation
On ce the license p late h as been
extracted , each ch aracter m ust b e
fragm ented .

Character Recognition
To iden tify characters, the seg men ted
characters in the licen se p an el m ust m atch
th e tem plates th at are already created .
PROGRESS OF THE PROJECT
OUTPUT
CONTINUE..
OUTPUT
 CONCLUSION

In conclusion, An Automatic Number Plate Recognition approach would basically extract the text/characters of the
number plate. For achieving this approach, we will be using techniques such as OpenCV, Image Processing, Optical
Character Recognition (OCR). Also, Automatic Number Plate Recognition systems would work fine in the dark and
low light environment.
 BIBLIOGRAPHY

• https://www.irjet.net/archives/V4/i6/IRJET-V4I6488. pdf

• https://www.analyticsvidhya.com/blog/2019/08/3-techniques-extract-features-from-image-data-machine-l
earning-python/#h-method-1-grayscale-pixel-values -as-features

• B. Pechiammal and J. A. Renjith, "An efficient approach for automatic license plate recognition system," 2017
Third International Conference on Science Technology Engineering & Management (ICONSTEM), 2017, pp.
121-129, doi: 10.1109/ICONSTEM.2017.8261267.
THANK YOU