Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor | Multi-disciplinary Studies | en_US |
dc.creator | Lee, Shu-tak Raymond | - |
dc.identifier.uri | https://theses.lib.polyu.edu.hk/handle/200/2999 | - |
dc.language | English | en_US |
dc.publisher | Hong Kong Polytechnic University | - |
dc.rights | All rights reserved | en_US |
dc.title | Invariant Chinese character recognition in dynamic link architecture | en_US |
dcterms.abstract | In this dissertation, a neural network model based on the application of "Dynamic Link" Architecture (DLA)" is presented for the off-line recognition of Chinese characters. "Dynamic Link Architecture(DLA)", a new generation of recurrent neural network model, first proposed by C. von der Malsbury in 1981, originated with the theory and mechanism of brain functions, with remarkable dynamic link memory features that can handle pattern recognition under various invariant transformation such as translation, rotation, dilation and distortion. In the report, a thorough overview on the methodology, network mechanism, and biological theory will be presented. From the implementation point of view, to tackle with the recognition of Chinese characters. A revised DLA model is used by using the 4-vector dynamic link assignment instead of the original 3-vector links (Wang, 1993). A selected set of Chinese characters will be tested for the performance of the recognition model under various transformation : translation, reflection, rotation, dilation and distortion (in certain extent). Challenging results are obtained. An improvement of 40% in recognition rate is attained by using the revised DLA model for Chinese character recognition. On the other hand, for the testing of invariant properties, an overall correct recognition rate of 85% is obtained under various transformations. In order to provide an overall picture for the contemporary character recognition models done by other co-workers, an overview of contemporary models and neural network architectures for character recognition, their basic methodologies, main features and performance on invariant properties will be discussed. | en_US |
dcterms.extent | ix, 119 leaves : ill. (some col.) ; 30 cm | en_US |
dcterms.isPartOf | PolyU Electronic Theses | en_US |
dcterms.issued | 1997 | en_US |
dcterms.educationalLevel | All Master | en_US |
dcterms.educationalLevel | M.Sc. | en_US |
dcterms.LCSH | Optical character recognition devices | en_US |
dcterms.LCSH | Chinese language -- Data processing | en_US |
dcterms.LCSH | Neural networks (Computer science) | en_US |
dcterms.LCSH | Hong Kong Polytechnic University -- Dissertations | en_US |
dcterms.accessRights | restricted access | en_US |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
b14054103.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 5.4 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/200/2999