This paper describes a Method of recognizing Hangeul (Korean Characters) which is based on the structural analysis of characters. The experiment results obtained through computer simulation are given. In recognition of Hangeul composed of 24 basic character elements, we pay special attention to regular combinational rules for character formation. The recognition system is composed of three major processing parts: Extraction of character elements, Recognition of each character element and Syntactic reconstruction of a character based on the recognized character elements. The character elements which failed in the syntactic reconstruction are sent back to the character element extraction process again, providing the flexibility to adopt to the more advanced experiments. We first present the structure of Hangeul for feature extraction. Next, some processing algorithms are described and simulated. The experiment results demonstrate the effectiveness of the proposed recognition system.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Byung-Uk CHOI, Tadao ICHIKAWA, Hiroichi FUJITA, "Recognition of Hangeul by Means of Structural Analysis" in IEICE TRANSACTIONS on transactions,
vol. E64-E, no. 5, pp. 319-325, May 1981, doi: .
Abstract: This paper describes a Method of recognizing Hangeul (Korean Characters) which is based on the structural analysis of characters. The experiment results obtained through computer simulation are given. In recognition of Hangeul composed of 24 basic character elements, we pay special attention to regular combinational rules for character formation. The recognition system is composed of three major processing parts: Extraction of character elements, Recognition of each character element and Syntactic reconstruction of a character based on the recognized character elements. The character elements which failed in the syntactic reconstruction are sent back to the character element extraction process again, providing the flexibility to adopt to the more advanced experiments. We first present the structure of Hangeul for feature extraction. Next, some processing algorithms are described and simulated. The experiment results demonstrate the effectiveness of the proposed recognition system.
URL: https://global.ieice.org/en_transactions/transactions/10.1587/e64-e_5_319/_p
Copy
@ARTICLE{e64-e_5_319,
author={Byung-Uk CHOI, Tadao ICHIKAWA, Hiroichi FUJITA, },
journal={IEICE TRANSACTIONS on transactions},
title={Recognition of Hangeul by Means of Structural Analysis},
year={1981},
volume={E64-E},
number={5},
pages={319-325},
abstract={This paper describes a Method of recognizing Hangeul (Korean Characters) which is based on the structural analysis of characters. The experiment results obtained through computer simulation are given. In recognition of Hangeul composed of 24 basic character elements, we pay special attention to regular combinational rules for character formation. The recognition system is composed of three major processing parts: Extraction of character elements, Recognition of each character element and Syntactic reconstruction of a character based on the recognized character elements. The character elements which failed in the syntactic reconstruction are sent back to the character element extraction process again, providing the flexibility to adopt to the more advanced experiments. We first present the structure of Hangeul for feature extraction. Next, some processing algorithms are described and simulated. The experiment results demonstrate the effectiveness of the proposed recognition system.},
keywords={},
doi={},
ISSN={},
month={May},}
Copy
TY - JOUR
TI - Recognition of Hangeul by Means of Structural Analysis
T2 - IEICE TRANSACTIONS on transactions
SP - 319
EP - 325
AU - Byung-Uk CHOI
AU - Tadao ICHIKAWA
AU - Hiroichi FUJITA
PY - 1981
DO -
JO - IEICE TRANSACTIONS on transactions
SN -
VL - E64-E
IS - 5
JA - IEICE TRANSACTIONS on transactions
Y1 - May 1981
AB - This paper describes a Method of recognizing Hangeul (Korean Characters) which is based on the structural analysis of characters. The experiment results obtained through computer simulation are given. In recognition of Hangeul composed of 24 basic character elements, we pay special attention to regular combinational rules for character formation. The recognition system is composed of three major processing parts: Extraction of character elements, Recognition of each character element and Syntactic reconstruction of a character based on the recognized character elements. The character elements which failed in the syntactic reconstruction are sent back to the character element extraction process again, providing the flexibility to adopt to the more advanced experiments. We first present the structure of Hangeul for feature extraction. Next, some processing algorithms are described and simulated. The experiment results demonstrate the effectiveness of the proposed recognition system.
ER -