Você está na página 1de 18

Project identification and

Literature survey-
survey-ECY 6197

License Plate Recognition System-


System-
License Plate identification

R.A.N.D.L.Rupasinghe
40464897
Aim
To develop a automated license plate recognition
system (LPR)

Objective
Identifying license plate from the extracted
license plate image
Vehicle identification?

 Why vehicle identification ?


 What are the methods already used for vehicle
identification ?
 What are the advantages disadvantages of
those methods ?
Why vehicle identification ?

 Vehicle identification is a valuable topic to


consider because there are lot of places that
needs this function

 Parking systems
 Traffic applications
 Security systems and ect
What are the methods already used for
vehicle identification ?

 Handheld electronic device


 Sequence of key strokes
 RFID tags
 Online access control systems
 Pulling wire
What is License Plate Recognition?

 license plate recognition (LPR) applies image


processing and character recognition technology
to identify vehicles by automatically reading their
license plates

Scope of the project


 In this case I only consider about identifying
vehicles which approaching to a certain gate
using those vehicles license plates
proposed system
Overview on license plate recognition
system and Elements of typical LPR
systems

Camera

Computer

various input/output boards


used to interface the
external world

Database
Main parts of the project

 Extraction of the license plate from the input


image
 Identifying the license plate from the extracted
image

 Identifying the license plate from the


extracted image
 Character segmentation
 Character identification
Character segmentation

 What is a character?
 Alphabetic characters (A,B,c,D,E)
 Numbers (1,2,5,7,.4)
 other characters (-
(-)

 Segmentation
 What is the segmentation ?
 What are the methods that can be use
What is the segmentation ?

 Image segmentation may be defined as the


partitioning of an image into regions or objects
that have coherent characteristics .

 Afetr that characters will extracted for further


processing such as recognition
Segmentation Methods

 Thresholding
Histogram Thresholding
Pixel Classification
 K-means Clustering
Region Growing
 Edge
Edge--based Approaches
Canny Edge Detection with Edge Linking
Thresholding

 Segmentation by Thresholding is based on the idea that:

 The pixels of segment in an image typically have similar


intensity values.
 Numerous pixels with the same intensity value should
be classified as their own segment.

 The general approach behind a thresholding algorithm is


as follows:

 A histogram of the image is created


 A threshold is selected.
 Anytime a histogram bar exceeds this threshold, that
is a new segment
Histogram Thresholding
Character identification

After segmentation of characters, the segmented


characters must identify before go further

 character recognition techniques that can


be use
 Statistical method
 Symbolic rule base method
 Harming distance approach
 Neural network approach
 What Is A Neural Network?
A computing system made up of a number of
simple, highly interconnected processing
elements, which process information by their
dynamic state response to external inputs.

 Types of neural net works


 Feed forward neural networks
 Radial basis function neural networks
 Kohenan self organizing Neural networks (self
organizing map)
 Recurrent net woks (RN)
Advantages of Neural Networks
 Adaptive learning
 Self
Self--Organization
 Real Time Operation
 Fault Tolerance
End

Você também pode gostar