Segmentation of merged characters by neural networks and shortest path
Identifieur interne : 000A50 ( PascalFrancis/Checkpoint ); précédent : 000A49; suivant : 000A51Segmentation of merged characters by neural networks and shortest path
Auteurs : JIN WANG [États-Unis] ; J. JeanSource :
- Pattern recognition [ 0031-3203 ] ; 1994.
Descripteurs français
- Pascal (Inist)
English descriptors
- KwdEn :
Abstract
A major problem with a neural network-based approach to printed character recognition is the segmentation of merged characters. A hybrid method is proposed which combines a neural network-based deferred segmentation scheme with conventional immediate segmentation techniques. In the deferred segmentation, a neural network is employed to distinguish single characters from composites. To find a proper vertical cut that separates a composite, a shortest-path algorithm seeking minimal-penalty curved cuts is used. Integrating those components with a multiresolution neural network OCR and an efficient spelling checker, the resulting system significantly improves its ability to read omnifont document text
Affiliations:
Links toward previous steps (curation, corpus...)
Links to Exploration step
Pascal:94-0585775Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">Segmentation of merged characters by neural networks and shortest path</title>
<author><name sortKey="Jin Wang" sort="Jin Wang" uniqKey="Jin Wang" last="Jin Wang">JIN WANG</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Wright State univ., dep. computer sci. eng.</s1>
<s2>Dayton OH 45435</s2>
<s3>USA</s3>
</inist:fA14>
<country>États-Unis</country>
<wicri:noRegion>Dayton OH 45435</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Jean, J" sort="Jean, J" uniqKey="Jean J" first="J." last="Jean">J. Jean</name>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">INIST</idno>
<idno type="inist">94-0585775</idno>
<date when="1994">1994</date>
<idno type="stanalyst">PASCAL 94-0585775 INIST</idno>
<idno type="RBID">Pascal:94-0585775</idno>
<idno type="wicri:Area/PascalFrancis/Corpus">000A98</idno>
<idno type="wicri:Area/PascalFrancis/Curation">000902</idno>
<idno type="wicri:Area/PascalFrancis/Checkpoint">000A50</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">Segmentation of merged characters by neural networks and shortest path</title>
<author><name sortKey="Jin Wang" sort="Jin Wang" uniqKey="Jin Wang" last="Jin Wang">JIN WANG</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Wright State univ., dep. computer sci. eng.</s1>
<s2>Dayton OH 45435</s2>
<s3>USA</s3>
</inist:fA14>
<country>États-Unis</country>
<wicri:noRegion>Dayton OH 45435</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Jean, J" sort="Jean, J" uniqKey="Jean J" first="J." last="Jean">J. Jean</name>
</author>
</analytic>
<series><title level="j" type="main">Pattern recognition</title>
<title level="j" type="abbreviated">Pattern recogn.</title>
<idno type="ISSN">0031-3203</idno>
<imprint><date when="1994">1994</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><title level="j" type="main">Pattern recognition</title>
<title level="j" type="abbreviated">Pattern recogn.</title>
<idno type="ISSN">0031-3203</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Character recognition</term>
<term>Document processing</term>
<term>Neural network</term>
<term>Pattern recognition</term>
<term>Segmentation</term>
<term>Shortest path</term>
</keywords>
<keywords scheme="Pascal" xml:lang="fr"><term>Chemin plus court</term>
<term>Segmentation</term>
<term>Réseau neuronal</term>
<term>Reconnaissance caractère</term>
<term>Reconnaissance forme</term>
<term>Traitement document</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">A major problem with a neural network-based approach to printed character recognition is the segmentation of merged characters. A hybrid method is proposed which combines a neural network-based deferred segmentation scheme with conventional immediate segmentation techniques. In the deferred segmentation, a neural network is employed to distinguish single characters from composites. To find a proper vertical cut that separates a composite, a shortest-path algorithm seeking minimal-penalty curved cuts is used. Integrating those components with a multiresolution neural network OCR and an efficient spelling checker, the resulting system significantly improves its ability to read omnifont document text</div>
</front>
</TEI>
<inist><standard h6="B"><pA><fA01 i1="01" i2="1"><s0>0031-3203</s0>
</fA01>
<fA02 i1="01"><s0>PTNRA8</s0>
</fA02>
<fA03 i2="1"><s0>Pattern recogn.</s0>
</fA03>
<fA05><s2>27</s2>
</fA05>
<fA06><s2>5</s2>
</fA06>
<fA08 i1="01" i2="1" l="ENG"><s1>Segmentation of merged characters by neural networks and shortest path</s1>
</fA08>
<fA11 i1="01" i2="1"><s1>JIN WANG</s1>
</fA11>
<fA11 i1="02" i2="1"><s1>JEAN (J.)</s1>
</fA11>
<fA14 i1="01"><s1>Wright State univ., dep. computer sci. eng.</s1>
<s2>Dayton OH 45435</s2>
<s3>USA</s3>
</fA14>
<fA20><s1>649-658</s1>
</fA20>
<fA21><s1>1994</s1>
</fA21>
<fA23 i1="01"><s0>ENG</s0>
</fA23>
<fA43 i1="01"><s1>INIST</s1>
<s2>15220</s2>
<s5>354000046073710030</s5>
</fA43>
<fA44><s0>0000</s0>
</fA44>
<fA45><s0>22 ref.</s0>
</fA45>
<fA47 i1="01" i2="1"><s0>94-0585775</s0>
</fA47>
<fA60><s1>P</s1>
</fA60>
<fA61><s0>A</s0>
</fA61>
<fA64 i1="01" i2="1"><s0>Pattern recognition</s0>
</fA64>
<fA66 i1="01"><s0>GBR</s0>
</fA66>
<fC01 i1="01" l="ENG"><s0>A major problem with a neural network-based approach to printed character recognition is the segmentation of merged characters. A hybrid method is proposed which combines a neural network-based deferred segmentation scheme with conventional immediate segmentation techniques. In the deferred segmentation, a neural network is employed to distinguish single characters from composites. To find a proper vertical cut that separates a composite, a shortest-path algorithm seeking minimal-penalty curved cuts is used. Integrating those components with a multiresolution neural network OCR and an efficient spelling checker, the resulting system significantly improves its ability to read omnifont document text</s0>
</fC01>
<fC02 i1="01" i2="X"><s0>001D02C03</s0>
</fC02>
<fC03 i1="01" i2="X" l="FRE"><s0>Chemin plus court</s0>
</fC03>
<fC03 i1="01" i2="X" l="ENG"><s0>Shortest path</s0>
</fC03>
<fC03 i1="01" i2="X" l="SPA"><s0>Camino más corto</s0>
</fC03>
<fC03 i1="02" i2="X" l="FRE"><s0>Segmentation</s0>
</fC03>
<fC03 i1="02" i2="X" l="ENG"><s0>Segmentation</s0>
</fC03>
<fC03 i1="02" i2="X" l="SPA"><s0>Segmentación</s0>
</fC03>
<fC03 i1="03" i2="X" l="FRE"><s0>Réseau neuronal</s0>
</fC03>
<fC03 i1="03" i2="X" l="ENG"><s0>Neural network</s0>
</fC03>
<fC03 i1="03" i2="X" l="SPA"><s0>Red neuronal</s0>
</fC03>
<fC03 i1="04" i2="X" l="FRE"><s0>Reconnaissance caractère</s0>
</fC03>
<fC03 i1="04" i2="X" l="ENG"><s0>Character recognition</s0>
</fC03>
<fC03 i1="04" i2="X" l="SPA"><s0>Reconocimiento carácter</s0>
</fC03>
<fC03 i1="05" i2="X" l="FRE"><s0>Reconnaissance forme</s0>
</fC03>
<fC03 i1="05" i2="X" l="ENG"><s0>Pattern recognition</s0>
</fC03>
<fC03 i1="05" i2="X" l="GER"><s0>Mustererkennung</s0>
</fC03>
<fC03 i1="05" i2="X" l="SPA"><s0>Reconocimiento patrón</s0>
</fC03>
<fC03 i1="06" i2="X" l="FRE"><s0>Traitement document</s0>
</fC03>
<fC03 i1="06" i2="X" l="ENG"><s0>Document processing</s0>
</fC03>
<fC03 i1="06" i2="X" l="SPA"><s0>Tratamiento documento</s0>
</fC03>
<fN21><s1>285</s1>
</fN21>
</pA>
</standard>
</inist>
<affiliations><list><country><li>États-Unis</li>
</country>
</list>
<tree><noCountry><name sortKey="Jean, J" sort="Jean, J" uniqKey="Jean J" first="J." last="Jean">J. Jean</name>
</noCountry>
<country name="États-Unis"><noRegion><name sortKey="Jin Wang" sort="Jin Wang" uniqKey="Jin Wang" last="Jin Wang">JIN WANG</name>
</noRegion>
</country>
</tree>
</affiliations>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/PascalFrancis/Checkpoint
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000A50 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/PascalFrancis/Checkpoint/biblio.hfd -nk 000A50 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Ticri/CIDE |area= OcrV1 |flux= PascalFrancis |étape= Checkpoint |type= RBID |clé= Pascal:94-0585775 |texte= Segmentation of merged characters by neural networks and shortest path }}
![]() | This area was generated with Dilib version V0.6.32. | ![]() |