An optimal approach towards recognizing broken Thai characters in OCR systems
This paper presents a novel technique for recognizing broken Thai characters found in degraded Thai text documents by modeling it as a set-partitioning problem (SPP). The technique searches for the optimal set-partition of the connected components by which each subset yields a reconstructed Thai cha...
Saved in:
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Conference or Workshop Item |
Published: |
2018
|
Subjects: | |
Online Access: | https://repository.li.mahidol.ac.th/handle/123456789/14005 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Mahidol University |
id |
th-mahidol.14005 |
---|---|
record_format |
dspace |
spelling |
th-mahidol.140052018-06-11T11:44:43Z An optimal approach towards recognizing broken Thai characters in OCR systems Chaivatna Sumetphong Supachai Tangwongsan Mahidol University Computer Science This paper presents a novel technique for recognizing broken Thai characters found in degraded Thai text documents by modeling it as a set-partitioning problem (SPP). The technique searches for the optimal set-partition of the connected components by which each subset yields a reconstructed Thai character. Given the non-linear nature of the objective function needed for optimal set-partitioning, we design an algorithm we call Heuristic Incremental Integer Programming (HIIP), that employs integer programming (IP) with an incremental approach using heuristics to hasten the convergence. To generate corrected Thai words, we adopt a probabilistic generative approach based a Thai dictionary corpus. The proposed technique is applied successfully to a Thai historical document and poor quality Thai fax document with promising accuracy rates over 93%. © 2012 IEEE. 2018-06-11T04:44:43Z 2018-06-11T04:44:43Z 2012-12-01 Conference Paper 2012 International Conference on Digital Image Computing Techniques and Applications, DICTA 2012. (2012) 10.1109/DICTA.2012.6411736 2-s2.0-84874352445 https://repository.li.mahidol.ac.th/handle/123456789/14005 Mahidol University SCOPUS https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84874352445&origin=inward |
institution |
Mahidol University |
building |
Mahidol University Library |
continent |
Asia |
country |
Thailand Thailand |
content_provider |
Mahidol University Library |
collection |
Mahidol University Institutional Repository |
topic |
Computer Science |
spellingShingle |
Computer Science Chaivatna Sumetphong Supachai Tangwongsan An optimal approach towards recognizing broken Thai characters in OCR systems |
description |
This paper presents a novel technique for recognizing broken Thai characters found in degraded Thai text documents by modeling it as a set-partitioning problem (SPP). The technique searches for the optimal set-partition of the connected components by which each subset yields a reconstructed Thai character. Given the non-linear nature of the objective function needed for optimal set-partitioning, we design an algorithm we call Heuristic Incremental Integer Programming (HIIP), that employs integer programming (IP) with an incremental approach using heuristics to hasten the convergence. To generate corrected Thai words, we adopt a probabilistic generative approach based a Thai dictionary corpus. The proposed technique is applied successfully to a Thai historical document and poor quality Thai fax document with promising accuracy rates over 93%. © 2012 IEEE. |
author2 |
Mahidol University |
author_facet |
Mahidol University Chaivatna Sumetphong Supachai Tangwongsan |
format |
Conference or Workshop Item |
author |
Chaivatna Sumetphong Supachai Tangwongsan |
author_sort |
Chaivatna Sumetphong |
title |
An optimal approach towards recognizing broken Thai characters in OCR systems |
title_short |
An optimal approach towards recognizing broken Thai characters in OCR systems |
title_full |
An optimal approach towards recognizing broken Thai characters in OCR systems |
title_fullStr |
An optimal approach towards recognizing broken Thai characters in OCR systems |
title_full_unstemmed |
An optimal approach towards recognizing broken Thai characters in OCR systems |
title_sort |
optimal approach towards recognizing broken thai characters in ocr systems |
publishDate |
2018 |
url |
https://repository.li.mahidol.ac.th/handle/123456789/14005 |
_version_ |
1763497664912031744 |