| Author: | Xu, Wendi |
| Title: | A comparative study of license plate localization methods with elevated video images |
| Degree: | M.Sc. |
| Year: | 2012 |
| Subject: | Automobile license plates -- China -- Hong Kong -- Identification. Image processing -- Data processing. Image processing -- Digital techniques. Hong Kong Polytechnic University -- Dissertations |
| Department: | Department of Land Surveying and Geo-Informatics |
| Pages: | x, 75 leaves : ill. (chiefly col.) ; 30 cm. |
| Language: | English |
| Abstract: | With the rapid growth in the number of vehicles in Hong Kong, it is necessary to develop a license plate recognition system to survey and monitor these vehicles. Since the precise license plate localization has a beneficial influence on the accuracy and the running speed of the license plate recognition system, choosing an appropriate localization method is important and necessary. This thesis presents a comparative study of license plate localization methods. A survey of the license plates in Hong Kong is conducted in this thesis. Several characteristics of the license plates and the vehicles are obtained from the survey. According to these characteristics, three localization methods are selected. These three localization methods are the color-based localization method, the mathematical morphology-based localization method and the frequency projection-based localization method. Numbers of vehicle images are processed by these three localization methods to locate the license plates. The results of the localizations are compared and analyzed in three aspects, which are the accuracy, the adaptability and the efficiency. After the comparison, the advantages and disadvantages of three localization methods are obtained. The color-based localization method has an ability of high running speed, and it also has drawbacks of bad accuracy and weak adaptability. The mathematical morphology-based localization method has good accuracy, and it also has bad efficiency and weak adaptability. The frequency projection-based localization method has advantages of high running speed and strong adaptability, and it also has a disadvantage of bad accuracy. Comparing three localization methods, the accuracy of the mathematical morphology-based localization method is the greatest and the accuracy of the frequency projection-based localization method is the lowest. In terms of adaptability, the frequency projection-based localization method has the strongest adaptability among three localization methods and the color-based localization method has the worst one. In terms of efficiency, the frequency projection-based localization method is the fastest method and the mathematical morphology-based localization method is the slowest one. According to the conclusions shown above, some proposals are given to implement an appropriate localization method for the license plates in Hong Kong. |
| Rights: | All rights reserved |
| Access: | restricted access |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| b24752897.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 3.32 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/6521