Tezislər / Theses


Supervisor: PhD, Associate Professor Leyla Muradkhanli



Yüklə 17,55 Mb.
Pdf görüntüsü
səhifə160/493
tarix02.10.2023
ölçüsü17,55 Mb.
#151572
1   ...   156   157   158   159   160   161   162   163   ...   493
BHOS Tezisler 2022 17x24sm

Supervisor: PhD, Associate Professor Leyla Muradkhanli
Keywords:
edge detection, segmentation, kernel, dilation, erosion
License Plate Recognition (LPR) system as known as Automatic 
Number Plate Recognition (ANPR) is the action of taking pictures of the 
license plates and transforming this optical data into the digital one by using 
image processing techniques in real-life time. The LPR systems consist of 


THE 3
rd
 INTERNATIONAL SCIENTIFIC CONFERENCES OF STUDENTS AND YOUNG RESEARCHERS 
dedicated to the 99
th
anniversary of the National Leader of Azerbaijan Heydar Aliyev
169
three standard phases which are number plate detection, plate character 
segmentation and character recognition. The detection of a number plate is 
a difficult operation due to wide variety of number plate formats and the fact 
that ambient variables cause problems during the image acquisition stage. 
Accurate character segmentation and recognition mainly rely on the effective 
plate detection. In order to locate and isolate the license plate and the 
characters, image processing techniques such as edge detection, 
thresholding, and resampling are used. LPR systems can be used to tackle 
several problems including vehicle identification, entrance admission to 
parking lots, security and airport cargo control, road traffic control, and radar 
systems.
The LPR system consists of two main parts which are hardware and 
software. The hardware part is the cameras which should combine Optical 
Character Recognition (OCR) and Neural Network in order to make the 
cameras to be able read the license plate of an object or a vehicle’s number 
plate and transform the data into the digital one in short time. The software 
behind the system can be written in different programming languages such 
as Python, C++ (with OpenCV library), MATLAB. The software consists of 
four main stages which are localization, segmentation, identification and 
regionalization. These stages are processed through the three steps: (I) Pre-
processing, (II) Localization, (III) Character segmentation and recognition. 
The first step includes three stages including converting images to 
grayscale, binarization and morphological image processing consisting of 
non-linear operations related to the shape or morphology of features of an 
image with the help of the structuring element as known as kernel. The main 
morphological process behind the software is dilation and erosion. The 
erosion process causes the image to shrink or decrease in the size; however, 
the dilation process makes the image dilate or grow in size. The shrink or 
increase size depends on the structuring element. Here are some examples 
of dilation and erosion (figure 1). 
Figure 1.
Dilation and erosion examples 
The localization of the license plate on an eroded image is one of the 
most crucial steps of the LPR system. This process can be done by using 
the image segmentation which is typically used in order to detect the objects. 
However, there can be some problems such as when the background color 



Yüklə 17,55 Mb.

Dostları ilə paylaş:
1   ...   156   157   158   159   160   161   162   163   ...   493




Verilənlər bazası müəlliflik hüququ ilə müdafiə olunur ©azkurs.org 2024
rəhbərliyinə müraciət

gir | qeydiyyatdan keç
    Ana səhifə


yükləyin