Development of automatic number plate recognition on android mobile phone platforms /
Automatic Number Plate Recognition (ANPR) is an intelligent system which has the capability to recognize the character on vehicle number plate. As the combination of hardware and software, ANPR is designed to offer the optimum reliability. Over the last decade, the researchers have proposed their me...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
Kuala Lumpur:
Kulliyyah of Engineering, International Islamic University Malaysia :
2014
|
| Subjects: | |
| Online Access: | Click here to view 1st 24 pages of the thesis. Members can view fulltext at the specified PCs in the library. |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Automatic Number Plate Recognition (ANPR) is an intelligent system which has the capability to recognize the character on vehicle number plate. As the combination of hardware and software, ANPR is designed to offer the optimum reliability. Over the last decade, the researchers have proposed their methods to recognize the vehicle number plate and implemented it in various access control, law enforcement and security, parking management system, toll gate or border access, tracking of stolen vehicles and traffic violations (speed trap). Most of the researchers implement the ANPR system on personal computer (PC) using high resolution camera and high computational capability. Only some of them have been designed and implemented the ANPR on mobile phone, which has limitation in camera resolution and processing speed. Prior to that, the main challenge of implementing ANPR algorithm on mobile phone is how to produce a higher coding efficiency with lower computational complexity and higher scalability. Hence, the objective of this research is to propose suitable and optimize algorithm for the development of ANPR system on Android mobile phone. In this thesis, various steps to optimize ANPR are described, such as image pre-processing, segmentation, and optical character recognition (OCR) using template matching. The proposed ANPR algorithm is based on an open source image processing library called Leptonica and OCR library called Tesseract. For comparison purposes, the template matching based OCR is compared to Artificial Neural Network (ANN) based OCR. Furthermore, the optimization on ANPR is performed on the image pre-processing step using our own Java code as currently there is no image processing library available on the standard Android mobile phone. Performance of the proposed algorithm is evaluated by the developed number plates' image database captured by mobile phone's camera, i.e. 30 images. Results showed that the accuracy and processing time of the proposed algorithm using template matching are 97.46% and 1.13 seconds, respectively. On the other hand, the traditional algorithm used template matching only obtained 83.65% accuracy with 0.97 second processing time. The result shows that our proposed algorithm has improved the accuracy of the ANPR with negligible additional processing time. |
|---|---|
| Physical Description: | xviii, 128 leaves : ill. ; 30cm. |
| Bibliography: | Includes bibliographical references (leaves 121-124). |