Structural primitive extraction and coding for handwritten numeral recognition
Identifieur interne : 002211 ( Main/Merge ); précédent : 002210; suivant : 002212Structural primitive extraction and coding for handwritten numeral recognition
Auteurs : Jianming Hu [Australie] ; Hong Yan [Australie]Source :
- Pattern Recognition [ 0031-3203 ] ; 1997.
Abstract
A structural method is proposed for unconstrained handwritten numeral recognition in this paper. The numeral is first smoothed and the skeleton is obtained. A set of feature points are then detected, and the skeleton is decomposed into primitives. A primitive code is defined to record the information of each primitive, and a global code is derived from the primitive codes to describe the topological structure of the skeleton. According to the global codes, all the numerals are classified into 26 subclasses. Two recognition algorithms have been developed based on the primitive codes. In the first algorithm, prototypes and matching rules are designed by hand. This allows a highly abstract matching rule being designed explicitly. Associated with each recognized numeral, a confidence level is also computed. In the second recognition algorithm, neural networks are used for each subclass, where the learning process can be carried out automatically. Good recognition results have been obtained with digit samples extracted from the NIST database. The performance of the recognition algorithms can still be improved if a more advanced thinning algorithm is used.
Url:
DOI: 10.1016/S0031-3203(97)00095-2
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: 000E78
- to stream Istex, to step Curation: 000E37
- to stream Istex, to step Checkpoint: 001598
Links to Exploration step
ISTEX:0CE070FA4BF59A9D1CD26F0B606C79BAF37302A6Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title>Structural primitive extraction and coding for handwritten numeral recognition</title>
<author><name sortKey="Hu, Jianming" sort="Hu, Jianming" uniqKey="Hu J" first="Jianming" last="Hu">Jianming Hu</name>
</author>
<author><name sortKey="Yan, Hong" sort="Yan, Hong" uniqKey="Yan H" first="Hong" last="Yan">Hong Yan</name>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">ISTEX</idno>
<idno type="RBID">ISTEX:0CE070FA4BF59A9D1CD26F0B606C79BAF37302A6</idno>
<date when="1998" year="1998">1998</date>
<idno type="doi">10.1016/S0031-3203(97)00095-2</idno>
<idno type="url">https://api.istex.fr/document/0CE070FA4BF59A9D1CD26F0B606C79BAF37302A6/fulltext/pdf</idno>
<idno type="wicri:Area/Istex/Corpus">000E78</idno>
<idno type="wicri:Area/Istex/Curation">000E37</idno>
<idno type="wicri:Area/Istex/Checkpoint">001598</idno>
<idno type="wicri:doubleKey">0031-3203:1998:Hu J:structural:primitive:extraction</idno>
<idno type="wicri:Area/Main/Merge">002211</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title level="a">Structural primitive extraction and coding for handwritten numeral recognition</title>
<author><name sortKey="Hu, Jianming" sort="Hu, Jianming" uniqKey="Hu J" first="Jianming" last="Hu">Jianming Hu</name>
<affiliation wicri:level="1"><country xml:lang="fr">Australie</country>
<wicri:regionArea>Department of Electrical Engineering, University of Sydney, NSW 2006</wicri:regionArea>
<wicri:noRegion>NSW 2006</wicri:noRegion>
</affiliation>
<affiliation wicri:level="1"><country wicri:rule="url">Australie</country>
</affiliation>
</author>
<author><name sortKey="Yan, Hong" sort="Yan, Hong" uniqKey="Yan H" first="Hong" last="Yan">Hong Yan</name>
<affiliation wicri:level="1"><country xml:lang="fr">Australie</country>
<wicri:regionArea>Department of Electrical Engineering, University of Sydney, NSW 2006</wicri:regionArea>
<wicri:noRegion>NSW 2006</wicri:noRegion>
</affiliation>
</author>
</analytic>
<monogr></monogr>
<series><title level="j">Pattern Recognition</title>
<title level="j" type="abbrev">PR</title>
<idno type="ISSN">0031-3203</idno>
<imprint><publisher>ELSEVIER</publisher>
<date type="published" when="1997">1997</date>
<biblScope unit="volume">31</biblScope>
<biblScope unit="issue">5</biblScope>
<biblScope unit="page" from="493">493</biblScope>
<biblScope unit="page" to="509">509</biblScope>
</imprint>
<idno type="ISSN">0031-3203</idno>
</series>
<idno type="istex">0CE070FA4BF59A9D1CD26F0B606C79BAF37302A6</idno>
<idno type="DOI">10.1016/S0031-3203(97)00095-2</idno>
<idno type="PII">S0031-3203(97)00095-2</idno>
</biblStruct>
</sourceDesc>
<seriesStmt><idno type="ISSN">0031-3203</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass></textClass>
<langUsage><language ident="en">en</language>
</langUsage>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">A structural method is proposed for unconstrained handwritten numeral recognition in this paper. The numeral is first smoothed and the skeleton is obtained. A set of feature points are then detected, and the skeleton is decomposed into primitives. A primitive code is defined to record the information of each primitive, and a global code is derived from the primitive codes to describe the topological structure of the skeleton. According to the global codes, all the numerals are classified into 26 subclasses. Two recognition algorithms have been developed based on the primitive codes. In the first algorithm, prototypes and matching rules are designed by hand. This allows a highly abstract matching rule being designed explicitly. Associated with each recognized numeral, a confidence level is also computed. In the second recognition algorithm, neural networks are used for each subclass, where the learning process can be carried out automatically. Good recognition results have been obtained with digit samples extracted from the NIST database. The performance of the recognition algorithms can still be improved if a more advanced thinning algorithm is used.</div>
</front>
</TEI>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Main/Merge
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 002211 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Merge/biblio.hfd -nk 002211 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Ticri/CIDE |area= OcrV1 |flux= Main |étape= Merge |type= RBID |clé= ISTEX:0CE070FA4BF59A9D1CD26F0B606C79BAF37302A6 |texte= Structural primitive extraction and coding for handwritten numeral recognition }}
This area was generated with Dilib version V0.6.32. |