Please use this identifier to cite or link to this item: http://ir.unikl.edu.my/jspui/handle/123456789/20627
metadata.theses.dc.title: MALAYSIAN LICENSE PLATE RECOGNITION SYSTEM BASED ON IMAGE PROCESSING USING SMEARING ALGORITHM
metadata.theses.dc.contributor.*: MUAYAD ALI HAMOOD BAKHTAN
metadata.theses.dc.date.issued: 28-Nov-2018
metadata.theses.dc.description.abstract: Plate Recognition becomes a significant tool in our daily life because of the unlimited increase of cars and transportation systems which made it impossible to be fully managed and monitored by humans. License Plate Recognition (LPR) system may contribute to solving by transportation problems. For example; minimizing the needs for guarding booths, increasing smooth traffic flow in front of parking gates, discovering vehicle frauds, or improving the efficiency of operational security in the premises. The system shall also benefit police departments to reduce traffic violations and track stolen vehicles. Although many researches have been conducted on license plates of different countries, license plate recognition in Malaysia is still an open research topic due to the distinct characteristics of Malaysian license plate numbers. This thesis aims to implement a technique of Malaysian License Plate Recognition (MLPR) based on smearing algorithm. The system operation begins with images capturing followed by plate localization, character segmentation, and ending with character recognition. At the end, the plate number is extracted from the image and used to determine the vehicle from a predefined database. In this research, 150 sample images are collected specifically for standard and non-standard Malaysian plates containing single row and double rows plates and based on specific effecting factors. Based on experimental results, the accuracy of license plate localization process is 97.4%, while the accuracy of the segmentation process and character recognition process are 96% and 76% respectively. The overall success rate of the proposed system reached 71% outperforming many of the existing techniques. Results also show that several environmental factors have significant impacts on the performance of plate recognition system.
metadata.theses.dc.identifier.uri: http://ir.unikl.edu.my/jspui/handle/123456789/20627
metadata.theses.dc.theses.semester: January 2017
metadata.theses.dc.theses.course: Degree of Master in Information Technology
Appears in Collections:Master Theses

Files in This Item:
File Description SizeFormat 
Master_Muayad Ali Hamood Bakhtan_MIIT.pdf80.1 kBAdobe PDFView/Open    Request a copy


Items in UniKL IR are protected by copyright, with all rights reserved, unless otherwise indicated.