Car plate recognition system is a system which aided human in
reading car plate number in an efficient way which saves energy and time. Before emerge of the computer-based car plate
recognition, human recorded car plate number manually by using pen and paper
until the introduction of the great technology in photo-enforcement industry. In the real world, there are also the car that
having the broken car plate. Therefore
the study will go further to ensure it also can be detected.
This problem should be solve because car plate recognition system
has a wide range of applications since the license plate of vehicles is the
most primary, widely accepted, human readable, mandatory identifier of motor
vehicles. This system has played significant
rules in traffic law enforcement and it is widely applied over the world to
identify stolen cars based on the up-to date blacklist. There are other useful applications such as
vehicle access control, automatic toll collection, real-time monitoring and
parking area security and management.
Plate number can be read by computer using proper step of image
processing which start from car plate localization, car plate extraction,
character segmentation and recognition of characters. However, when no standard font style and size
is maintained in different Vehicle Number Plates, identificationof characters
in a Number Plate becomes difficult .Detection of location of the Number Plate
from image is also difficult task. All
these problems restrict to the most efficience method and algorithm used in
order to detect a car plate.
Futhermore, this study also will be used for futher reaserch
purposes in order to study how to calculate the accuaracy of possible letter or
number due to the broken character on a car plate. This future research will be conduct if this
paper achive the objective listed and relevance for a long time. This future reasearch will help to recognize
and determine the broken character on car plate and then it will very helpful
to the Investigation Industries to solve any kind of cases that involving the
detection of broken car plate. Thus,
this current research are very important and usefull to expand the knowledge
about car plate detection and develop the study about it that can be use to
solve any related problems occur in the real world related to car plate
detection and recognisation.
This chapter is divided into ten chapters. Chapter 2 provides a description of the
problem statements while chapter 3 defines the objectives of the project and
chapter 4 defines the significance of the project. In chapter 5, the scope of project is
discussed and chapter 6 will explain the unusual definition of term and concept. Next, literature review will be deliberating
in chapter 7 and methodology for this study will be enlightened in chapter 8. Finally, project benefit and project
scheduling are being stated in chapter 9 and chapter 10 respectively.
This study of car plate detection sometimes brings to face some
difficulties and boundries in order to meet successcion especially in Malaysia. This is beause of existance of cars that does
not follow the standardize car plate that has been specified by the gorvement. Besides that.
the method used like Canny and Mathematical Morphological be a big
challenge to satisfied the perfect and smooth image detection. for example
by using canny method, there exist some gap between the boundries of the
image because of noise. Canny would be a
lot of broken high-detail area edge for images with complex textures and noisy,
and it is difficult to form a large area (Zhang,2010). Next by using Mathematical Morphological,
there certain part of the image can not be detected perfectly because of weak
The objectives of this project are:
1) To identify the car plate
images using canny and mathematical morphological.
2) To compare the effectiveness
of canny and mathematical morphological or the combination of both
3) To purpose the guideline for
identification of car plate images to be used by the authority body.
SIGNIFICANCE OF THE PROJECT
The finding of this project will
help the Investigation Instituition to solve any kind of cases that involving
detection of car plate wheter its in good condition or broken. Besides, it can be used by Road Transport
Department (JPJ) to detecet the road user that againts traffic laws. The detection appilication that used by JPJ
can be seen in the example of Automated Enforcement System (AES) that act as a
speed trap to capture images of traffic offenders. Other than that, this finding will help
security agencies to get the clearer images detection by using Closed-Circuit
Television (CCTV) as their security devices to record or capture the image of
criminal’s car plate in their area.
5. SCOPE OF THE PROJECT
In order to carrying out this
project, there are several boundries and limitation that should be considered. To investigate the broken letter or number on
the plate number, first of all, the plate number must follow the standardized Malaysia
plate number that has been specified by government. Secondly, the broken plate number that can be
considered in this project must at least still have 50% of its original letter
or number. If the plate numbers are
broke for more than 50% or fully broken, it cannot be taken as a sample in this
6. DEFINITION OF TERMS AND CONCEPTS
Canny : Canny edge detection is a image
processing method used to detect edges in an image while suppressing noise.
Mathematical morphological : Mathematical morphology is a technique used
not only on images, but also on graphs, meshes, and anything else that has a
spatial nature to it.
RGB (red, green, and blue) refers to a system for
representing the colors to be used on a computer display. Red, green, and blue can be combined in
various proportions to obtain any color in the visible spectrum. Levels of R, G, and B can each range from 0
to 100 percent of full intensity (Rouse, 2010).
Dilation on image :Dilation is the process of improvising given image by filling holes
in a image, sharpen the edges of objects in an image, and join the broken lines
and increase the brightness of an image (Varsha and Pawan, 2014).