Serveur d'exploration sur l'OCR

Attention, ce site est en cours de développement !
Attention, site généré par des moyens informatiques à partir de corpus bruts.
Les informations ne sont donc pas validées.

Handwritten Chinese characters recognition by greedy matching with geometric constraint

Identifieur interne : 002667 ( Main/Exploration ); précédent : 002666; suivant : 002668

Handwritten Chinese characters recognition by greedy matching with geometric constraint

Auteurs : Ai-Jia Hsieh [République populaire de Chine] ; Kuo-Chin Fan [République populaire de Chine, Taïwan] ; Tzu-I Fan [République populaire de Chine]

Source :

RBID : ISTEX:8A4B9FAE99BE672986FA6F1005B86F090A61F6E3

Abstract

In this paper, a greedy matching algorithm with geometric constraint for solving the polygonal arcs matching (boundary line segments matching) is proposed. Assume the matched cost and the unmatched penalty of each line segment are given, as well as any two matched pairs in the matching which preserves the geometric relation. Our goal is to find a matching such that (1) the sum of costs of matched line segments and the penalties of unmatched line segments is minimum, and (2) the matching preserves the geometric relation. The proposed greedy matching algorithm consists of two main modules. The first is an optimal matching module, which utilizes the optimization matching method to find an optimal matching without the geometric constraint. The second module is an evaluation and update module, which deletes the matched pair with a geometric relation having the maximum inconsistency after each matched pair has been evaluated, and a new matching is found again. The algorithm continues until a stable matching is found. To verify the validity of the proposed algorithm, we implement it to recognize handwritten Chinese postal characters. We selected 51 Chinese postal characters as the prototype characters, and the experiments are conducted on 2550 samples, with each category containing 50 variations. The recognition rate is 91.8%. Experimental results reveal the feasibility of our proposed method in recognizing handwritten Chinese postal characters.

Url:
DOI: 10.1016/0262-8856(95)01043-2


Affiliations:


Links toward previous steps (curation, corpus...)


Le document en format XML

<record>
<TEI wicri:istexFullTextTei="biblStruct">
<teiHeader>
<fileDesc>
<titleStmt>
<title>Handwritten Chinese characters recognition by greedy matching with geometric constraint</title>
<author>
<name sortKey="Hsieh, Ai Jia" sort="Hsieh, Ai Jia" uniqKey="Hsieh A" first="Ai-Jia" last="Hsieh">Ai-Jia Hsieh</name>
</author>
<author>
<name sortKey="Fan, Kuo Chin" sort="Fan, Kuo Chin" uniqKey="Fan K" first="Kuo-Chin" last="Fan">Kuo-Chin Fan</name>
</author>
<author>
<name sortKey="Fan, Tzu I" sort="Fan, Tzu I" uniqKey="Fan T" first="Tzu-I" last="Fan">Tzu-I Fan</name>
</author>
</titleStmt>
<publicationStmt>
<idno type="wicri:source">ISTEX</idno>
<idno type="RBID">ISTEX:8A4B9FAE99BE672986FA6F1005B86F090A61F6E3</idno>
<date when="1996" year="1996">1996</date>
<idno type="doi">10.1016/0262-8856(95)01043-2</idno>
<idno type="url">https://api.istex.fr/document/8A4B9FAE99BE672986FA6F1005B86F090A61F6E3/fulltext/pdf</idno>
<idno type="wicri:Area/Istex/Corpus">000424</idno>
<idno type="wicri:Area/Istex/Curation">000417</idno>
<idno type="wicri:Area/Istex/Checkpoint">001A78</idno>
<idno type="wicri:doubleKey">0262-8856:1996:Hsieh A:handwritten:chinese:characters</idno>
<idno type="wicri:Area/Main/Merge">002811</idno>
<idno type="wicri:Area/Main/Curation">002667</idno>
<idno type="wicri:Area/Main/Exploration">002667</idno>
</publicationStmt>
<sourceDesc>
<biblStruct>
<analytic>
<title level="a">Handwritten Chinese characters recognition by greedy matching with geometric constraint</title>
<author>
<name sortKey="Hsieh, Ai Jia" sort="Hsieh, Ai Jia" uniqKey="Hsieh A" first="Ai-Jia" last="Hsieh">Ai-Jia Hsieh</name>
<affiliation wicri:level="1">
<country xml:lang="fr" wicri:curation="lc">République populaire de Chine</country>
<wicri:regionArea>Institute of Computer Science and Information Engineering, National Central University, Chungli, Taiwan 32054</wicri:regionArea>
<wicri:noRegion>Taiwan 32054</wicri:noRegion>
</affiliation>
<affiliation wicri:level="1">
<country xml:lang="fr" wicri:curation="lc">République populaire de Chine</country>
<wicri:regionArea>Computer & Communication Research Laboratories, Industrial Technology Research Institute, Hsinchu, Taiwan 31015</wicri:regionArea>
<wicri:noRegion>Taiwan 31015</wicri:noRegion>
</affiliation>
</author>
<author>
<name sortKey="Fan, Kuo Chin" sort="Fan, Kuo Chin" uniqKey="Fan K" first="Kuo-Chin" last="Fan">Kuo-Chin Fan</name>
<affiliation>
<wicri:noCountry code="no comma">Corresponding author.</wicri:noCountry>
</affiliation>
<affiliation wicri:level="1">
<country xml:lang="fr" wicri:curation="lc">République populaire de Chine</country>
<wicri:regionArea>Institute of Computer Science and Information Engineering, National Central University, Chungli, Taiwan 32054</wicri:regionArea>
<wicri:noRegion>Taiwan 32054</wicri:noRegion>
</affiliation>
<affiliation wicri:level="1">
<country wicri:rule="url">Taïwan</country>
</affiliation>
</author>
<author>
<name sortKey="Fan, Tzu I" sort="Fan, Tzu I" uniqKey="Fan T" first="Tzu-I" last="Fan">Tzu-I Fan</name>
<affiliation wicri:level="1">
<country xml:lang="fr" wicri:curation="lc">République populaire de Chine</country>
<wicri:regionArea>Institute of Computer Science and Information Engineering, National Central University, Chungli, Taiwan 32054</wicri:regionArea>
<wicri:noRegion>Taiwan 32054</wicri:noRegion>
</affiliation>
</author>
</analytic>
<monogr></monogr>
<series>
<title level="j">Image and Vision Computing</title>
<title level="j" type="abbrev">IMAVIS</title>
<idno type="ISSN">0262-8856</idno>
<imprint>
<publisher>ELSEVIER</publisher>
<date type="published" when="1994">1994</date>
<biblScope unit="volume">14</biblScope>
<biblScope unit="issue">2</biblScope>
<biblScope unit="page" from="91">91</biblScope>
<biblScope unit="page" to="104">104</biblScope>
</imprint>
<idno type="ISSN">0262-8856</idno>
</series>
<idno type="istex">8A4B9FAE99BE672986FA6F1005B86F090A61F6E3</idno>
<idno type="DOI">10.1016/0262-8856(95)01043-2</idno>
<idno type="PII">0262-8856(95)01043-2</idno>
</biblStruct>
</sourceDesc>
<seriesStmt>
<idno type="ISSN">0262-8856</idno>
</seriesStmt>
</fileDesc>
<profileDesc>
<textClass></textClass>
<langUsage>
<language ident="en">en</language>
</langUsage>
</profileDesc>
</teiHeader>
<front>
<div type="abstract" xml:lang="en">In this paper, a greedy matching algorithm with geometric constraint for solving the polygonal arcs matching (boundary line segments matching) is proposed. Assume the matched cost and the unmatched penalty of each line segment are given, as well as any two matched pairs in the matching which preserves the geometric relation. Our goal is to find a matching such that (1) the sum of costs of matched line segments and the penalties of unmatched line segments is minimum, and (2) the matching preserves the geometric relation. The proposed greedy matching algorithm consists of two main modules. The first is an optimal matching module, which utilizes the optimization matching method to find an optimal matching without the geometric constraint. The second module is an evaluation and update module, which deletes the matched pair with a geometric relation having the maximum inconsistency after each matched pair has been evaluated, and a new matching is found again. The algorithm continues until a stable matching is found. To verify the validity of the proposed algorithm, we implement it to recognize handwritten Chinese postal characters. We selected 51 Chinese postal characters as the prototype characters, and the experiments are conducted on 2550 samples, with each category containing 50 variations. The recognition rate is 91.8%. Experimental results reveal the feasibility of our proposed method in recognizing handwritten Chinese postal characters.</div>
</front>
</TEI>
<affiliations>
<list>
<country>
<li>République populaire de Chine</li>
<li>Taïwan</li>
</country>
</list>
<tree>
<country name="République populaire de Chine">
<noRegion>
<name sortKey="Hsieh, Ai Jia" sort="Hsieh, Ai Jia" uniqKey="Hsieh A" first="Ai-Jia" last="Hsieh">Ai-Jia Hsieh</name>
</noRegion>
<name sortKey="Fan, Kuo Chin" sort="Fan, Kuo Chin" uniqKey="Fan K" first="Kuo-Chin" last="Fan">Kuo-Chin Fan</name>
<name sortKey="Fan, Tzu I" sort="Fan, Tzu I" uniqKey="Fan T" first="Tzu-I" last="Fan">Tzu-I Fan</name>
<name sortKey="Hsieh, Ai Jia" sort="Hsieh, Ai Jia" uniqKey="Hsieh A" first="Ai-Jia" last="Hsieh">Ai-Jia Hsieh</name>
</country>
<country name="Taïwan">
<noRegion>
<name sortKey="Fan, Kuo Chin" sort="Fan, Kuo Chin" uniqKey="Fan K" first="Kuo-Chin" last="Fan">Kuo-Chin Fan</name>
</noRegion>
</country>
</tree>
</affiliations>
</record>

Pour manipuler ce document sous Unix (Dilib)

EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 002667 | SxmlIndent | more

Ou

HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 002667 | SxmlIndent | more

Pour mettre un lien sur cette page dans le réseau Wicri

{{Explor lien
   |wiki=    Ticri/CIDE
   |area=    OcrV1
   |flux=    Main
   |étape=   Exploration
   |type=    RBID
   |clé=     ISTEX:8A4B9FAE99BE672986FA6F1005B86F090A61F6E3
   |texte=   Handwritten Chinese characters recognition by greedy matching with geometric constraint
}}

Wicri

This area was generated with Dilib version V0.6.32.
Data generation: Sat Nov 11 16:53:45 2017. Site generation: Mon Mar 11 23:15:16 2024