EzTran: A Mobile-based Chinese Character Recognition and Translation Service System
Date Issued
2010
Date
2010
Author(s)
Chen, I-Chun
Abstract
As the popularity of smart phone and wireless technologies, people are getting used to search for information from the Internet anytime and everywhere. In addition, since the resolution of embedded cameras on mobile devices becomes higher and most smart phones are equipped with the touch screen, a large number of interesting and useful applications have been proposed to improve human life. It is observed that although the translation function can be commonly seen on mobile phones, it is very difficult for foreigners to type Chinese characters for searching. Moreover, for native users, it is still time-consuming and inconvenient to type Chinese characters on portable devices. To solve these problems, we propose a Chinese character recognition and translation service system, EzTran, which is a practical system that provides users to get the translated result in English and the related web pages searched from Google search engine easily by only snapping the text photo.
Moreover, EzTran can also be integrated with other applications. In this thesis, we demonstrate the implementation of a food recommendation system with a shop sign recognition scheme. Extensive experiments have been conducted to verify the effectiveness of the proposed system. Experimental results show that EzTran can generate translation outputs in real time with satisfactory recognition accuracy.
Subjects
pattern recognition
mobile-based system
Chinese character recognition
computer vision
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-99-R97921025-1.pdf
Size
23.32 KB
Format
Adobe PDF
Checksum
(MD5):5d5c4fa49ddc887f9cb59d55332d27fd