A character decomposition approach to offline handwritten Chinese character recognition

Pao Yue-kong Library Electronic Theses Database

A character decomposition approach to offline handwritten Chinese character recognition

 

Author: Ip, Wai-shun Wilson
Title: A character decomposition approach to offline handwritten Chinese character recognition
Degree: M.Phil.
Year: 1998
Subject: Optical character recognition devices
Chinese character sets (Data processing)
Chinese characters -- Data processing
Optical pattern recognition
Pattern recognition systems
Optical data processing
Hong Kong Polytechnic University -- Dissertations
Department: Dept. of Computing
Pages: vii, 86 leaves : ill. ; 30 cm
Language: English
InnoPac Record: http://library.polyu.edu.hk/record=b1446449
URI: http://theses.lib.polyu.edu.hk/handle/200/810
Abstract: Optical handwritten Chinese character recognition is regarded as a difficult and challenging problem. The major difficulties encountered in solving this problem are the large number of character classes, the high complexity and mutual similarity of Chinese characters, and the large variations in writing styles. In general, the recognition problem can be divided into two categories, on-line and off-line. In on-line recognition, a light pen is used to write characters on a tablet so that the strokes can be extracted orderly according to the writing sequences. Unlike on-line one, the input mechanism of off-line recognition does not offer information like positions, sequences and types of strokes. Only a 2-D character image is available for processing. Furthermore, human interaction during the recognition process is usually not available and hence the performance requirement of such system should be much higher. Since Chinese characters are composed of radicals, it is apparent to recognize radicals inside a character before recognizing the character because matching the simplified components is much easier and faster than matching the whole complex character. Despite also the fact that Chinese usually formulate their knowledge of Chinese characters as a combination of radicals, very few studies have focused on this character decomposition approach to recognition. In this research, we have adopted such an approach and the problem of how to extract radical sub-images from character images is particularly addressed. A radical extraction algorithm based on deformable templates (DTs) has been developed. Deformable templates generally possess shape-varying capability, making them particularly suitable for extracting and recognizing non-rigid objects. In fact, they have been successfully applied to problems like contour detection, motion tracking, and object matching. The application of DTs to Chinese character recognition is a novel one and concepts like goodness of character decomposition have been exploited to formulate appropriate energy terms and to devise cost effective minimization schemes for the problem. An energy function is defined to pull the decomposition template away from the strokes of a character. The template then interacts dynamically with the image to minimize the energy function, thereby deforming itself to find the best decomposition. The advantage of the character decomposition approach is demonstrated by feeding the extracted radical images to an adopted structural based Chinese character recognizer whose outputs are then combined to produce the class label of the input character. Simulation results show that the performance of the adopted Chinese character recognition system can be improved significantly when character decomposition approach is used.

Files in this item

Files Size Format
b14464494.pdf 3.335Mb PDF
Copyright Undertaking
As a bona fide Library user, I declare that:
  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. 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.
  3. 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.

     

Quick Search

Browse

More Information