Recognition of large-set printed Hangul (Korean script) by two-stage backpropagation neural classifier

Sung-Bae Cho, Jin H. Kim

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

A two-stage neural network classifier is described which practically recognizes printed Hangul (Korean script). This classifier is composed of a type classification network and six recognition networks. The former classifies input character images into one of the six types by their overall structure, and then the latter classify them into character code. Furthermore, a training scheme including systematic noises is introduced for improving the generalization capability of the networks. Experiments are conducted with the most frequently used 990 printed Hangul syllables. By the noise included training, the recognition rate amounts to 98.28%, which is better than that of the conventional backpropagation learning. A comparison with a statistical classifier and an analysis of generalization capability confirm the relative superiority of the proposed classification method.

Original languageEnglish
Pages (from-to)1353-1360
Number of pages8
JournalPattern Recognition
Volume25
Issue number11
DOIs
Publication statusPublished - 1992 Jan 1

Fingerprint

Backpropagation
Classifiers
Neural networks
Experiments

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

@article{4d3fd618cd3d4fee9560bcbf8e61530f,
title = "Recognition of large-set printed Hangul (Korean script) by two-stage backpropagation neural classifier",
abstract = "A two-stage neural network classifier is described which practically recognizes printed Hangul (Korean script). This classifier is composed of a type classification network and six recognition networks. The former classifies input character images into one of the six types by their overall structure, and then the latter classify them into character code. Furthermore, a training scheme including systematic noises is introduced for improving the generalization capability of the networks. Experiments are conducted with the most frequently used 990 printed Hangul syllables. By the noise included training, the recognition rate amounts to 98.28{\%}, which is better than that of the conventional backpropagation learning. A comparison with a statistical classifier and an analysis of generalization capability confirm the relative superiority of the proposed classification method.",
author = "Sung-Bae Cho and Kim, {Jin H.}",
year = "1992",
month = "1",
day = "1",
doi = "10.1016/0031-3203(92)90147-B",
language = "English",
volume = "25",
pages = "1353--1360",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "11",

}

Recognition of large-set printed Hangul (Korean script) by two-stage backpropagation neural classifier. / Cho, Sung-Bae; Kim, Jin H.

In: Pattern Recognition, Vol. 25, No. 11, 01.01.1992, p. 1353-1360.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Recognition of large-set printed Hangul (Korean script) by two-stage backpropagation neural classifier

AU - Cho, Sung-Bae

AU - Kim, Jin H.

PY - 1992/1/1

Y1 - 1992/1/1

N2 - A two-stage neural network classifier is described which practically recognizes printed Hangul (Korean script). This classifier is composed of a type classification network and six recognition networks. The former classifies input character images into one of the six types by their overall structure, and then the latter classify them into character code. Furthermore, a training scheme including systematic noises is introduced for improving the generalization capability of the networks. Experiments are conducted with the most frequently used 990 printed Hangul syllables. By the noise included training, the recognition rate amounts to 98.28%, which is better than that of the conventional backpropagation learning. A comparison with a statistical classifier and an analysis of generalization capability confirm the relative superiority of the proposed classification method.

AB - A two-stage neural network classifier is described which practically recognizes printed Hangul (Korean script). This classifier is composed of a type classification network and six recognition networks. The former classifies input character images into one of the six types by their overall structure, and then the latter classify them into character code. Furthermore, a training scheme including systematic noises is introduced for improving the generalization capability of the networks. Experiments are conducted with the most frequently used 990 printed Hangul syllables. By the noise included training, the recognition rate amounts to 98.28%, which is better than that of the conventional backpropagation learning. A comparison with a statistical classifier and an analysis of generalization capability confirm the relative superiority of the proposed classification method.

UR - http://www.scopus.com/inward/record.url?scp=0026946349&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0026946349&partnerID=8YFLogxK

U2 - 10.1016/0031-3203(92)90147-B

DO - 10.1016/0031-3203(92)90147-B

M3 - Article

AN - SCOPUS:0026946349

VL - 25

SP - 1353

EP - 1360

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 11

ER -