school

UM E-Theses Collection (澳門大學電子學位論文庫)

check Full Text
Title

License Plate Recognition algorithms and their application to Macao license plates

English Abstract

Macao has been developing quickly in the decade and the amount of vehicles gets increasing rapidly. Many problems arise and the need of an advanced Intelligent Transport Solution is obvious. License Plate Recognition (LPR) is one of the most important topics for this. In this research, the core problems of LPR, their challenges and general approaches will be discussed. In our work, an edge-based method is proposed to locate the license plate. Sobel edge detector is employed to extract the vertical edges, and an edge composition algorithm is proposed to combine the edges into candidate plate regions. They are further examined on the existence of the character “M” by a verification algorithm with template matching. It is a Macao country specific feature that license plate number must contain the character “M” as the first character. A row separation algorithm with projection analysis is also proposed to cater both one-row and two-row types of plates. With the plate region identified, projection analysis and template matching methods are exploited to segment and recognize the characters. Various pre and post processing steps such as morphological erosion and trimming are proposed other than traditional implementation so as to improve the recognition accuracy.

Issue date

2010

Author

Ho, Wai Yiu

Faculty

Faculty of Science and Technology

Degree

M.Sc.

Subject

Automobile license plates

Automobile license plates -- Macau

Software Engineering -- Department of Computer and Information Science

Supervisor

Pun, Chi Man

Files In This Item

Full-text (Intranet)

Full-text (Intranet)

Location
1/F Zone C
Library URL
991005548139706306