1

ARTIFICIAL INTELLIGENCE
JURY PRESENTATION

TOPIC: AI IN NUMBER PLATE RECOGNITION

Submitted to -Jury Panel

Submitted by -Vibhav Sharnagat

Yashaswini Raj
 Artificial Inteligence
The modern definition of artificial intelligence (or AI) is
"the study and design of intelligent agents" where an
intelligent agent is a system that perceives its
environment and takes actions which maximizes its
chances of success.
John mccarthy, who coined the term in 1956, defines it
as "the science and engineering of making intelligent
machines.“
Othernames for the field have been proposed, such
as computational intelligence, synthetic intelligence or
computational rationality.
The term artificial intelligence is also used to describe a
property of machines or programs: the intelligence that
the system demonstrates.
 INTRODUCTION
Automatic Number Plate Recognition 3

ANPR is a technology that uses optical character

recognition on images to read vehicle registration plates to
create vehicle location data.

It can use existing closed-circuit television, road-rule

enforcement cameras, or cameras specifically designed for
the task.

ANPR is used by police forces to check if a vehicle is

registered or licensed.

It is also used for electronic toll collection on pay-per-

use roads and as a method of cataloguing the
movements of traffic, for example by highways agencies.
 4
The developed system first detects the vehicle and
then captures the vehicle image.

 Vehiclenumber plate region is extracted using the

image segmentation in an image.

Optical character recognition technique is used for the

character recognition.

The resulting data is then used to compare with the

records on a database so as to come up with the
specific information like the vehicle owner, place of
registration, address, etc.
 Working-Existing System
5
The ANPR process is divided into the following three steps:-

1. Detection of the vehicle:-

 The first step is to take an image of the vehicle at

the right time.

 The ANPR equipment controls physically a sensor

directly installed in the lane.

 Whenever a vehicle has been detected by the

sensor, the ANPR equipment will know its presence,
and then the process of the capture begins.
 2. Capturing Of Image 6

 The ANPR equipment does not need to receive signal

from any external sensor.

 The ANPR equipment takes images continuously and it

is able to detect the vehicles automatically Capture of
the images.

 Once the vehicle is detected, the following step is the

capture of the vehicle. In order to take a right image.

 The ANPR unit activates the illumination and takes

picturesof the front and rear plates using ANPR
camera.
 7

3. Process of Recognition

 The ANPR unit analyzes the image with

different image processing software.

 The ANPR unit checks if the vehicle appears on

a predefined list of authorized cars.

 The system waits for the next vehicle to

approach the secured area.
The work is divided into several parts: 8

1. Input car image:- Input the image that is

taken of the car.

2. Gray scale conversion:- From the input RGB

image it has to be converted to gray scale and the
8-bit grey value calculated.
3.
•
Reduce noise using mid-filtering method:-
It basically uses median filtering technique to reduce
9
the paper and salt noise.
• It used 3x 3 masks to get eight neighbors of a pixel
and their corresponding gray value.

4. License Plate Detection:-

• The basic step in recognition of vehicle number
plate is to detect the plate size.
• In general number plates are rectangular in shape.
Hence we have to detect the edges of the
rectangular plate.
• Mathematical morphology will be used to detect
that region.
• Using Sobel edge detector we used to high light
regions with a high edge magnitude and high edge
variance are identified.
5. Segmentation of characters:-
• The next step is segmentation of the license 10
plate area into smaller parts which represent
each character of the license plate
• This is done using the vertical projection.

6. Character Recognition:- The process of character

recognition is repeated for each character image
obtained in the last step.
•The output of this process should be a recognized
character.
•The outputs are, letters of the alphabet, numbers from 0
to 9 and special characters like a dash.
•The initial step is to separate possible outputs into
smaller groups counting the character end points
•For example, character 'I' has 2 end points, but on the
other hand, character '9' has only 1 end points.
This character recognition allows us to use a 3×3 mask to find one of 8 possible types of
end points. Every 3×3 area of an image is compared to 8 masks 11
 PROBLEMS WITH THE EXISTING SYSTEM
12

 The images of the number plate or of any object which is taken by using the optical
character reader technology may get blurred mainly due to the reason of motion blurring
for which the picture seems to be hazy when uploaded in the database.

 The technology often uses low-resolution images for which the images are not actually visible
properly in every case.

 The License plate structure is not uniform as stipulated by the

laws of Indian Government. Majority vehicle owners do not
comply with the norms foresaid for the Size and color of the
license plate.
 PROBLEMS WITH THE EXISTING SYSTEM (Cont..) 13

 The number plate is not recognized properly if there is

use of multiple lines to write the number.

 Vehicle lane change in the camera's angle of

view during license plate reading.

 The technology often got images with disturbed light

intensity by which the images are not actually visible
properly in every case.
 Solution 14
 A standard rule for all over the country has to implemented for writing the systematic
number or the registration plates such as Style, Font, Color, Proportion of every letter
because by fancy style number it is very hard to identify the character by image
recognition.

 Character line number – to determine whether characters are spanned in more than one line
or column. By using algorithm it can identify the number is in how many line this algorithm is
not in current ANPR system after license plate detection this algorithm has to be input after
that image segmentation should be done.
 Solution (Count..) 15
 Light management the type of light is as important as how it is managed. It is possible to use
any typical technique of photography, we can use two ways to get proper light for the
image.

• To control the light emitted by the focus.

• To control the light that enters in the camera.
 Solution Through A.I 16

Use Of ANN

 Artificial Neural Network (ANN) sometimes known as neural network is a mathematical

term, which contains interconnected artificial neurons.
 In1024 X 768 image size ratio two layer probabilistic neural network.
 The character recognition process was performed in 128ms.
 In 220x50 image size ratio multi layered perceptron (MLP) ANN model is used for
classification of characters.
 In contains input layer for decision making, hidden layer to compute more complicated
associations and output layer for resulting decision.
 Feed forward back-propagation (BP) algorithm was used to train ANN.
 HNN is applied to reduce ambiguity between the similar characters e.g. 8 and B, 2 and Z
etc.
 The authors claim to have more than 99% recognition rate.
 Solution Through A.I 17
Template Matching

 Template matching is useful for recognition of fixed sized characters.

 It can be also used for detection of objects generally in face detection and medical image processing.
 It is further divided in two parts: feature based matching and template based matching. Feature based
approach is useful when template image has strong features otherwise template based approach can
be useful.
 In statistical feature extraction method is applied for achieving 85% of character recognition rate.
 In 640 X 480 image size ratio several features and extracted and salient is computed based on training
characters.
 A linear normalization algorithm is used to adjust all characters with uniform size. The recognition rate
of 95.7% is achieved among 1176 images.
 It used low-resolution template matching method to work with lower resolution image such as 4 X 8.
 The similarity function is used to measure similarity between patterns.
 INPUT and OUTPUT 18

Identification of
Greyscale Contrast
vehicle in a Noise removal
Conversion Enhancement
lane

Actual Number Image

Image
Plate Area Edge Detection Subtraction
Binarization
Extraction Operation

Display of the
Extracted Plate Character Character
Vehicle
Region Segmentation Recognition
Number
ADVANTAGES OF THE SYSTEM
 Improving Road Safety. 19

 Reduces Crime; Recovering stolen cars

 Gives Officers Better Information To Work With.

 Deterring Terrorism.

 Giving A Greater Police Presence.

 Increasing The Percentage Of Stop/Searches

That Lead To An Arrest.

 Pre-paid Parking Members Can Be Easily

Differentiated From Non-members.
 APPLICATIONS
20

Parking

Traffic Access-
Control Control

Applications
of ANPR

Stolen
Tolling
Cars

Border
Control
 PARKING
 The plate number is used to automatically enter
pre-paid members and calculate parking fee for
non-members (by comparing the exit and entry
times).
ACCESS-CONTROL
21
A gate automatically opens for authorized
members in a secured area, thus replacing or
assisting the security guard.
The events are logged on a database and could be
used to search the history of events.
 TOLLING BORDER CONTROL 22
 The car number is used to calculate the travel  This installation covers the borders of the entire

fee in a toll- road, or used to double-check Country.

the ticket.  Each vehicle is registered into a central database and

linked to additional information such as the passport
data. This is used to track all border crossings.
 CONCLUSION 23

 There is an immediate need of such kind of Automatic Number Plate Recognition

system in India as there are problems of traffic, stealing cars etc.

 Government should take some interest in developing this system as this system
is very economical and eco-friendly, if applied effectively .

 This change will help in the progress of the nation.

 Reference 24

 https://www.researchgate.net/publication/236888959_Automatic_Nu
mber_Plate_Recognition_System_ANPR_A_Survey/download

 https://www.netsolutions.com/insights/using-apples-machine-learning-
for-license-plate-recognition/

 https://medium.com/nurture-ai/ravi-theja-implementing-ai-research-
paper-for-the-first-time-46852170ccdc

 https://www.irjet.net/archives/V4/i3/IRJET-V4I3242.pdf
 25

Thank You!