1. GOVERNMENT ENGINEERING COLLEGE RAMANAGARA
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
Phase 2 Project Review
on
“Number Plate Recognition for Automatic No-Parking with Zero
Tolerance”
Project code:18CSP77
PRESENTED BY
Pallavi B S [1GG20CS416]
Gagan R M [1GG20CS405]
Arkesha J [1GG20CS401]
Divya A V [1GG19CS013]
UNDER THE GUIDANCE OF
Dr. SHABEEN TAJ G A
Assistant Professor of Computer
Science and Engineering Department
3. ABSTRACT
• Traffic law violation has been recognized as a major cause for road accidents in
most parts of the world with majority occurring in developing countries
• Therefore, a system needs to be designed to assist law enforcement agencies to
impose these rules to improve road safety and reduce road accidents.
• This work uses Automatic No-parking detection and Vehicle Plate Number
Recognition (VPNR) which is a real-time embedded system to automatically
recognize license plate numbers.
• The main aim of the system is to use image processing to identify vehicles
violating traffic rules by their plate numbers. The system if implemented can be
used to improve road safety and control traffic of emerging smart cities.
4. INTRODUCTION
• In the current era of information technology, the use of automatic system is
becoming more and more widespread.
• Due to automation the human labour as been replacing by automated machines.
• Generally in the no parking zones they inform drivers and motorist were parking
not permitted.
• Vehicle Number Plate Recognition (VNPR) is one of the most effective tools
currently available.
• Vehicle plate number recognition(VNPR) system was invented by the police
scientific development branch from the UK.
5. EXISTING SYSTEM
• In existing system traffic police will identify vehicle
and charge penalty.
• Traffic police will capture number plate of that vehicle
and send notice to the owner.
• In the existing system is very accurate but not efficient
to monitor all the vehicle on the NO PARKING ZONE.
• After surveying all these difficulties we proposed the
advent of new technology to avoid traffic problem.
6. PROBLEM STATEMENT
• Traffic congestion has become one of the major problems in the Indian traffic
system.
• Vehicles in the no parking zones leads to the narrowing of the roadway and thus
leads to the occurrence of the traffic congestion.
• Frequently park in the no parking zones that may results in violation of law.
• Traditional monitoring and security systems where manpower used is slower.
7. PROPOSED SYSTEM
• The proposed system is built on a Raspberry Pi with a camera to
capture a no parking scene.
• Ultrasonic sensor which detects objects within its field of vision.
8. PROPOSED SYSTEM
• The Raspberry Pi extracts the number plate part of the
image.
• Extract the numbers from the number plate and compare
with database.
• Then an email will send with details of traffic violation
to traffic department and owner.
9. LITERATURE SURVEY
YEAR TITLE AUTHOR ABSTRACT DISADVANTAGE
2018 IOT Based No-parking
Notifier System(IEEE)
Sanjith. M. Gowda G
Sushanth S Sujatha
In future, this system can be used
to deduce the tolls and parking
charges directly form the vehicle
owner’s online wallet.
To monitor the vehicle on
the road it require RFID
and Cloud system all
time is not possible.
2018 Automatic Number Plate
Recognition
Abhishek Kashyap, B.
Suresh, Anukul Patil
Number Plate Recognition system
is asecuritysystem. Image
processing concept is used
inNumber Plate Recognition
system OCR (OpticalCharacter
Recognition) scheme is also
applied in this forreading the
image of vehicle number plate.
ID's with colored
background can be
problematic to OCR
10. Year Title Author Abstract Disadvantage
2020 A Novel Method for Indian Vehicle
Registration Number Plate Detection
and Recognition using Image
Processing Techniques(General)
Ravi Kiran
Varma,Srikanth
Gantaa, Hari
Krishna B,Praveen
The device which uses
deskewing and k
nearest neighbour
algorithm for character
recognition
The device which can not
process detection and
recognition into a single
framework.
2022 Raspberry pi and opencv for license
plate recognition in
realtime(General)
Simeen S.
Mujawar,Sreerekh
a Vadi
The built system
correctly detects the
car number plate
region in the picture.
This system is not efficient to
give accurate output.
11. OBJECTIVE
• The proposed system is used to detect illegal parked vehicles and this
system used for imposed to fine on vehicle.
• The proposed technique will reduce the need of man power for
monitoring and security applications.
• This proposed framework will be helping to avoid the traffic
congestions
15. IMPLEMENTATION
EasyOCR module
EasyOCR is actually a python package that holds PyTorch.
EasyOCR like any other OCR(tesseract of Google or any other) detects the text
from images
Image preprocessing:The input can either be an image or a video. Video is
considered as a series of Images/frames, before starting with Number plate
detection, the Image source must be made suitable for further processing.
License Plate Detection :To detect the number or license plate we need to find the
contours on the image.
Character Segmentation: Segmentation is nothing but breaking the whole image
into subparts to process them further.
Character Recognition: Optical character recognition(OCR), is a process of
recognizing text inside images and converting it into an electronic form
18. SYSTEM SPECIFICATION
Hardware Requirements
• Raspberry pi : module 3B
• Camara : PI Camara 5 MP
• Processor : Minimum 1.6 GHz or Faster performance.
• RAM : Minimum 1GB or more.
• ROM : Minimum 10GB or more.
• CPU Cores : Minimum 1 core.
20. TESTING
Test Case ID Test Case Title Description Expected Output Result
1 Working raspberry pi model
Mack a connection and check the all the connections as
per the over requirements. Project started at the Raspberry pi.
PASS
2
Start Raspberry pi Project Write the python sample code on raspberry pi and run the
code
Output is displayed PASS
3
Working of camara in raspberry pi
recommender
Make the camara connection and run the code of sample
camara enabling
Working of camara PASS
4 Write a code for pi Camara
Run the source doe.
Check if any error gives while importing the necessary
packages
Project running successfully PASS
21. 5 Adding sensor in the raspberry pi
Check the connection of all the pins of sensor are
connected rightly and with the GRO AND ERCP
Connection successfully PASS
6
Write a code for the sensor to
trigger and camara captured the
image
Run the source code and check it working condition is
correctly are not
working Successful PASS
7 Firebase connectivity
The input should be in the specified format and comes
the raspberry pi picamara . Correct input should be accepted PASS
8 Optical characters recognition
Run the code of OCR and The output should be different
for all the different inputs
The number plates are deferent
Different outputs should be given for
different inputs
PASS
9
SMS communication
And fast2sms connectivity
Create a accountin Fast2SMS and Run the code and the . working Successful PASS
10 Flow directory The project should follow a predefined path.
The flow of the project should be
represented in the high-level design.
PASS
11 Handling errors If the server stops responding. 404 error page should be displayed PASS
22. CONCLUSIONS AND FUTURE ENHANCEMENTS
The proposed system will eliminate need of human beings for monitoring and security
applications.
This system will not require physical presence of human at no parking area to take action
against illegally parked vehicles.
This system facilitates authority to take action against owner of illegally parked vehicles.
This simple image processing approach can be used for different application with constant
background such as: (i) Automated toll collection, (ii) Access control, (iii) Border security
etc.
Some of the difficulties in recognition of number plates: (i) Broken number plate, (ii)
similarity between certain characters (0 and D, 5 and S, 8 and B etc.) (iii) Number plate
not within the legal specifications, (iv) Plate partially visible or dirt on the plate etc.