Distinctive features in automatic recognition of handprinted characters
Identifieur interne : 000333 ( Istex/Corpus ); précédent : 000332; suivant : 000334Distinctive features in automatic recognition of handprinted characters
Auteurs : Ching Y. SuenSource :
- Signal Processing [ 0165-1684 ] ; 1981.
Abstract
Characters are recognized from the features extracted. Usually the input chacter is smoothed and cleaned by the preprocessor before it reaches the featuree extractor. Good preprocessors and feature extractors are the prerequisites to a successful character recognition system. Following a description of preprocessing techniques, the various features found in the vast accumulation of literature on handprint recognition are divided into two main categories: (1) global analysis, and (2) structural analysis. Further subdivision of these categories gives rise to six families of feature type, viz. (a) distribution of points, (b) transformation, (c) physical measurements, (d) line segments and edges, (e) outline of character, and (f) centre-line of character. Each family is described with illustrative examples. The performance and recognition rates of systems employing these features are discussed.
Url:
DOI: 10.1016/0165-1684(82)90021-4
Links to Exploration step
ISTEX:72E9014115E3F1E2F0F9B12B3FA520D242150C90Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title>Distinctive features in automatic recognition of handprinted characters</title><author><name sortKey="Suen, Ching Y" sort="Suen, Ching Y" uniqKey="Suen C" first="Ching Y." last="Suen">Ching Y. Suen</name><affiliation><mods:affiliation>Department of Computer Science, Concordia University, Montreal, Canada H3G 1M8</mods:affiliation></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:72E9014115E3F1E2F0F9B12B3FA520D242150C90</idno><date when="1982" year="1982">1982</date><idno type="doi">10.1016/0165-1684(82)90021-4</idno><idno type="url">https://api.istex.fr/document/72E9014115E3F1E2F0F9B12B3FA520D242150C90/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">000333</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a">Distinctive features in automatic recognition of handprinted characters</title><author><name sortKey="Suen, Ching Y" sort="Suen, Ching Y" uniqKey="Suen C" first="Ching Y." last="Suen">Ching Y. Suen</name><affiliation><mods:affiliation>Department of Computer Science, Concordia University, Montreal, Canada H3G 1M8</mods:affiliation></affiliation></author></analytic><monogr></monogr><series><title level="j">Signal Processing</title><title level="j" type="abbrev">SIGPRO</title><idno type="ISSN">0165-1684</idno><imprint><publisher>ELSEVIER</publisher><date type="published" when="1981">1981</date><biblScope unit="volume">4</biblScope><biblScope unit="issue">2–3</biblScope><biblScope unit="page" from="193">193</biblScope><biblScope unit="page" to="207">207</biblScope></imprint><idno type="ISSN">0165-1684</idno></series><idno type="istex">72E9014115E3F1E2F0F9B12B3FA520D242150C90</idno><idno type="DOI">10.1016/0165-1684(82)90021-4</idno><idno type="PII">0165-1684(82)90021-4</idno></biblStruct></sourceDesc><seriesStmt><idno type="ISSN">0165-1684</idno></seriesStmt></fileDesc><profileDesc><textClass></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Characters are recognized from the features extracted. Usually the input chacter is smoothed and cleaned by the preprocessor before it reaches the featuree extractor. Good preprocessors and feature extractors are the prerequisites to a successful character recognition system. Following a description of preprocessing techniques, the various features found in the vast accumulation of literature on handprint recognition are divided into two main categories: (1) global analysis, and (2) structural analysis. Further subdivision of these categories gives rise to six families of feature type, viz. (a) distribution of points, (b) transformation, (c) physical measurements, (d) line segments and edges, (e) outline of character, and (f) centre-line of character. Each family is described with illustrative examples. The performance and recognition rates of systems employing these features are discussed.</div></front></TEI><istex><corpusName>elsevier</corpusName><author><json:item><name>Ching Y. Suen</name><affiliations><json:string>Department of Computer Science, Concordia University, Montreal, Canada H3G 1M8</json:string></affiliations></json:item></author><subject><json:item><lang><json:string>eng</json:string></lang><value>Feature extraction</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>handprint recognition</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>recognition algorithms</value></json:item></subject><language><json:string>eng</json:string></language><qualityIndicators><score>6.512</score><pdfVersion>1.2</pdfVersion><pdfPageSize>540 x 756 pts</pdfPageSize><refBibsNative>true</refBibsNative><keywordCount>3</keywordCount><abstractCharCount>903</abstractCharCount><pdfWordCount>7153</pdfWordCount><pdfCharCount>44467</pdfCharCount><pdfPageCount>15</pdfPageCount><abstractWordCount>126</abstractWordCount></qualityIndicators><title>Distinctive features in automatic recognition of handprinted characters</title><pii><json:string>0165-1684(82)90021-4</json:string></pii><genre><json:string>research-article</json:string></genre><serie><pages><last>232</last><first>225</first></pages><genre></genre><language><json:string>unknown</json:string></language><title>Proc. EJCC</title></serie><host><volume>4</volume><pii><json:string>S0165-1684(00)X0293-9</json:string></pii><pages><last>207</last><first>193</first></pages><issn><json:string>0165-1684</json:string></issn><issue>2–3</issue><genre><json:string>Journal</json:string></genre><language><json:string>unknown</json:string></language><title>Signal Processing</title><publicationDate>1982</publicationDate></host><categories><wos><json:string>ENGINEERING, ELECTRICAL & ELECTRONIC</json:string></wos></categories><publicationDate>1981</publicationDate><copyrightDate>1982</copyrightDate><doi><json:string>10.1016/0165-1684(82)90021-4</json:string></doi><id>72E9014115E3F1E2F0F9B12B3FA520D242150C90</id><fulltext><json:item><original>true</original><mimetype>application/pdf</mimetype><extension>pdf</extension><uri>https://api.istex.fr/document/72E9014115E3F1E2F0F9B12B3FA520D242150C90/fulltext/pdf</uri></json:item><json:item><original>true</original><mimetype>text/plain</mimetype><extension>txt</extension><uri>https://api.istex.fr/document/72E9014115E3F1E2F0F9B12B3FA520D242150C90/fulltext/txt</uri></json:item><json:item><original>false</original><mimetype>application/zip</mimetype><extension>zip</extension><uri>https://api.istex.fr/document/72E9014115E3F1E2F0F9B12B3FA520D242150C90/fulltext/zip</uri></json:item><istex:fulltextTEI uri="https://api.istex.fr/document/72E9014115E3F1E2F0F9B12B3FA520D242150C90/fulltext/tei"><teiHeader><fileDesc><titleStmt><title level="a">Distinctive features in automatic recognition of handprinted characters</title></titleStmt><publicationStmt><authority>ISTEX</authority><publisher>ELSEVIER</publisher><availability><p>ELSEVIER</p></availability><date>1982</date></publicationStmt><notesStmt><note type="content">Section title: Regular paper</note></notesStmt><sourceDesc><biblStruct type="inbook"><analytic><title level="a">Distinctive features in automatic recognition of handprinted characters</title><author><persName><forename type="first">Ching Y.</forename><surname>Suen</surname></persName><note type="biography">member EURASIP</note><affiliation>member EURASIP</affiliation><affiliation>Department of Computer Science, Concordia University, Montreal, Canada H3G 1M8</affiliation></author></analytic><monogr><title level="j">Signal Processing</title><title level="j" type="abbrev">SIGPRO</title><idno type="pISSN">0165-1684</idno><idno type="PII">S0165-1684(00)X0293-9</idno><imprint><publisher>ELSEVIER</publisher><date type="published" when="1981"></date><biblScope unit="volume">4</biblScope><biblScope unit="issue">2–3</biblScope><biblScope unit="page" from="193">193</biblScope><biblScope unit="page" to="207">207</biblScope></imprint></monogr><idno type="istex">72E9014115E3F1E2F0F9B12B3FA520D242150C90</idno><idno type="DOI">10.1016/0165-1684(82)90021-4</idno><idno type="PII">0165-1684(82)90021-4</idno></biblStruct></sourceDesc></fileDesc><profileDesc><creation><date>1982</date></creation><langUsage><language ident="en">en</language></langUsage><abstract xml:lang="en"><p>Characters are recognized from the features extracted. Usually the input chacter is smoothed and cleaned by the preprocessor before it reaches the featuree extractor. Good preprocessors and feature extractors are the prerequisites to a successful character recognition system. Following a description of preprocessing techniques, the various features found in the vast accumulation of literature on handprint recognition are divided into two main categories: (1) global analysis, and (2) structural analysis. Further subdivision of these categories gives rise to six families of feature type, viz. (a) distribution of points, (b) transformation, (c) physical measurements, (d) line segments and edges, (e) outline of character, and (f) centre-line of character. Each family is described with illustrative examples. The performance and recognition rates of systems employing these features are discussed.</p></abstract><abstract xml:lang="de"><p>Zeichen werden aufgrund von Merkmalen erkannt. Üblicherweise wird einzu erkennedes Zeichen in der Vorverarbeitungsstufe geglättet und ‘gesäubert’, bevor es den Merkmalsextraktor erreicht. Gute Vorverarbeitungsstufen und Merkmalsextraktoren sind hierbei die Voraussetzung für erfolgreiches Arbeiten eines Zeichenerkennungssystems. In diesem Beitrag werden zunächst einige Prinzipien der Vorverarbeitung beschrieben; hierauf werden die verschiedenen Verarbeitungsmethoden, die sich in dem umfangreichen Schrifttum über die automatische Erkennung handschriftlicher Zeichen finden, in zwei Hauptkategorien eingeteilt: (1) globale Analyse sowie (2) strukturelle Analyse. Eine weitere Unterteilung dieser Kategorien ergibt je nach Verarbeitungsmethode sechs Merkmalsfamilien: (a) Verteilung der Zeichenpunkte, (b) Transformation, (c) physikalische Messungen, (d) Liniensegmente und Ränder, (e) Umriβ sowie (f) Skelett des Zeichens. Jede dieser Merkmalsfamilien wird beschrieben und anhand von Beispielen erklärt. Abschlieβend wird die Wirkungsweise und die Erkennungsrate von Erkennungssystemen diskutiert, die diese Merkmale verwenden.</p></abstract><abstract xml:lang="fr"><p>On reconnait les caractères à partir des attibuts extraits. En général, le caractère d'entrée estt adouci et nettoyé par le préprocesseur avant qu'il atteigne l'extracteur d'attribut. De bons préprocesseurs et extracteurs d'attribut sont nécessaires à un système de reconnaissance de caractère performant. Après une description de techniques de préprocesseurs, les différents attibuts toruvés dans la grande quantité de littérature traitant de la reconnaissance de caractère manuscrits sont divisés en deux catégories principales: (1) analyse global et (2) analyse de structure. Une subdivision de ces catégories donne lieu à six familles de types d'attribut, à savoir: (a) distribution de points, (b) transformation, (c) mesures physiques, (d) segments de droite et contours, (e) profil du caractère et (f) ligne centrale du caractère. Chaque famille est décrite à l'aide d'exemples illustratifs. Les performances et les taux de reconnaissance des systèmes utilisant ces attributs sont discutés.</p></abstract><textClass><keywords scheme="keyword"><list><head>Keywords</head><item><term>Feature extraction</term></item><item><term>handprint recognition</term></item><item><term>recognition algorithms</term></item></list></keywords></textClass></profileDesc><revisionDesc><change when="1981-01-20">Received</change><change when="1981-06-24">Modified</change><change when="1981">Published</change></revisionDesc></teiHeader></istex:fulltextTEI></fulltext><metadata><istex:metadataXml wicri:clean="Elsevier, elements deleted: tail"><istex:xmlDeclaration>version="1.0" encoding="utf-8"</istex:xmlDeclaration><istex:docType PUBLIC="-//ES//DTD journal article DTD version 4.5.2//EN//XML" URI="art452.dtd" name="istex:docType"></istex:docType><istex:document><converted-article version="4.5.2" docsubtype="fla"><item-info><jid>SIGPRO</jid><aid>82900214</aid><ce:pii>0165-1684(82)90021-4</ce:pii><ce:doi>10.1016/0165-1684(82)90021-4</ce:doi><ce:copyright type="unknown" year="1982"></ce:copyright></item-info><head><ce:dochead><ce:textfn>Regular paper</ce:textfn></ce:dochead><ce:title>Distinctive features in automatic recognition of handprinted characters</ce:title><ce:author-group><ce:author><ce:given-name>Ching Y.</ce:given-name><ce:surname>Suen</ce:surname><ce:roles>member EURASIP</ce:roles></ce:author><ce:affiliation><ce:textfn>Department of Computer Science, Concordia University, Montreal, Canada H3G 1M8</ce:textfn></ce:affiliation></ce:author-group><ce:date-received day="20" month="1" year="1981"></ce:date-received><ce:date-revised day="24" month="6" year="1981"></ce:date-revised><ce:date-revised day="8" month="12" year="1981"></ce:date-revised><ce:abstract><ce:section-title>Abstract</ce:section-title><ce:abstract-sec><ce:simple-para>Characters are recognized from the features extracted. Usually the input chacter is smoothed and cleaned by the preprocessor before it reaches the featuree extractor. Good preprocessors and feature extractors are the prerequisites to a successful character recognition system. Following a description of preprocessing techniques, the various features found in the vast accumulation of literature on handprint recognition are divided into two main categories: (1) global analysis, and (2) structural analysis. Further subdivision of these categories gives rise to six families of feature type, viz. (a) distribution of points, (b) transformation, (c) physical measurements, (d) line segments and edges, (e) outline of character, and (f) centre-line of character. Each family is described with illustrative examples. The performance and recognition rates of systems employing these features are discussed.</ce:simple-para></ce:abstract-sec></ce:abstract><ce:abstract xml:lang="de"><ce:section-title>Zusammenfassung</ce:section-title><ce:abstract-sec><ce:simple-para>Zeichen werden aufgrund von Merkmalen erkannt. Üblicherweise wird einzu erkennedes Zeichen in der Vorverarbeitungsstufe geglättet und ‘gesäubert’, bevor es den Merkmalsextraktor erreicht. Gute Vorverarbeitungsstufen und Merkmalsextraktoren sind hierbei die Voraussetzung für erfolgreiches Arbeiten eines Zeichenerkennungssystems. In diesem Beitrag werden zunächst einige Prinzipien der Vorverarbeitung beschrieben; hierauf werden die verschiedenen Verarbeitungsmethoden, die sich in dem umfangreichen Schrifttum über die automatische Erkennung handschriftlicher Zeichen finden, in zwei Hauptkategorien eingeteilt: (1) globale Analyse sowie (2) strukturelle Analyse. Eine weitere Unterteilung dieser Kategorien ergibt je nach Verarbeitungsmethode sechs Merkmalsfamilien: (a) Verteilung der Zeichenpunkte, (b) Transformation, (c) physikalische Messungen, (d) Liniensegmente und Ränder, (e) Umriβ sowie (f) Skelett des Zeichens. Jede dieser Merkmalsfamilien wird beschrieben und anhand von Beispielen erklärt. Abschlieβend wird die Wirkungsweise und die Erkennungsrate von Erkennungssystemen diskutiert, die diese Merkmale verwenden.</ce:simple-para></ce:abstract-sec></ce:abstract><ce:abstract xml:lang="fr"><ce:section-title>Résumé</ce:section-title><ce:abstract-sec><ce:simple-para>On reconnait les caractères à partir des attibuts extraits. En général, le caractère d'entrée estt adouci et nettoyé par le préprocesseur avant qu'il atteigne l'extracteur d'attribut. De bons préprocesseurs et extracteurs d'attribut sont nécessaires à un système de reconnaissance de caractère performant. Après une description de techniques de préprocesseurs, les différents attibuts toruvés dans la grande quantité de littérature traitant de la reconnaissance de caractère manuscrits sont divisés en deux catégories principales: (1) analyse global et (2) analyse de structure. Une subdivision de ces catégories donne lieu à six familles de types d'attribut, à savoir: (a) distribution de points, (b) transformation, (c) mesures physiques, (d) segments de droite et contours, (e) profil du caractère et (f) ligne centrale du caractère. Chaque famille est décrite à l'aide d'exemples illustratifs. Les performances et les taux de reconnaissance des systèmes utilisant ces attributs sont discutés.</ce:simple-para></ce:abstract-sec></ce:abstract><ce:keywords><ce:section-title>Keywords</ce:section-title><ce:keyword><ce:text>Feature extraction</ce:text></ce:keyword><ce:keyword><ce:text>handprint recognition</ce:text></ce:keyword><ce:keyword><ce:text>recognition algorithms</ce:text></ce:keyword></ce:keywords></head></converted-article></istex:document></istex:metadataXml><mods version="3.6"><titleInfo><title>Distinctive features in automatic recognition of handprinted characters</title></titleInfo><titleInfo type="alternative" contentType="CDATA"><title>Distinctive features in automatic recognition of handprinted characters</title></titleInfo><name type="personal"><namePart type="given">Ching Y.</namePart><namePart type="family">Suen</namePart><affiliation>Department of Computer Science, Concordia University, Montreal, Canada H3G 1M8</affiliation><description>member EURASIP</description><role><roleTerm type="text">author</roleTerm></role></name><typeOfResource>text</typeOfResource><genre type="research-article" displayLabel="Full-length article"></genre><originInfo><publisher>ELSEVIER</publisher><dateIssued encoding="w3cdtf">1981</dateIssued><dateCaptured encoding="w3cdtf">1981-01-20</dateCaptured><dateModified encoding="w3cdtf">1981-06-24</dateModified><copyrightDate encoding="w3cdtf">1982</copyrightDate></originInfo><language><languageTerm type="code" authority="iso639-2b">eng</languageTerm><languageTerm type="code" authority="rfc3066">en</languageTerm></language><physicalDescription><internetMediaType>text/html</internetMediaType></physicalDescription><abstract lang="en">Characters are recognized from the features extracted. Usually the input chacter is smoothed and cleaned by the preprocessor before it reaches the featuree extractor. Good preprocessors and feature extractors are the prerequisites to a successful character recognition system. Following a description of preprocessing techniques, the various features found in the vast accumulation of literature on handprint recognition are divided into two main categories: (1) global analysis, and (2) structural analysis. Further subdivision of these categories gives rise to six families of feature type, viz. (a) distribution of points, (b) transformation, (c) physical measurements, (d) line segments and edges, (e) outline of character, and (f) centre-line of character. Each family is described with illustrative examples. The performance and recognition rates of systems employing these features are discussed.</abstract><abstract lang="de">Zeichen werden aufgrund von Merkmalen erkannt. Üblicherweise wird einzu erkennedes Zeichen in der Vorverarbeitungsstufe geglättet und ‘gesäubert’, bevor es den Merkmalsextraktor erreicht. Gute Vorverarbeitungsstufen und Merkmalsextraktoren sind hierbei die Voraussetzung für erfolgreiches Arbeiten eines Zeichenerkennungssystems. In diesem Beitrag werden zunächst einige Prinzipien der Vorverarbeitung beschrieben; hierauf werden die verschiedenen Verarbeitungsmethoden, die sich in dem umfangreichen Schrifttum über die automatische Erkennung handschriftlicher Zeichen finden, in zwei Hauptkategorien eingeteilt: (1) globale Analyse sowie (2) strukturelle Analyse. Eine weitere Unterteilung dieser Kategorien ergibt je nach Verarbeitungsmethode sechs Merkmalsfamilien: (a) Verteilung der Zeichenpunkte, (b) Transformation, (c) physikalische Messungen, (d) Liniensegmente und Ränder, (e) Umriβ sowie (f) Skelett des Zeichens. Jede dieser Merkmalsfamilien wird beschrieben und anhand von Beispielen erklärt. Abschlieβend wird die Wirkungsweise und die Erkennungsrate von Erkennungssystemen diskutiert, die diese Merkmale verwenden.</abstract><abstract lang="fr">On reconnait les caractères à partir des attibuts extraits. En général, le caractère d'entrée estt adouci et nettoyé par le préprocesseur avant qu'il atteigne l'extracteur d'attribut. De bons préprocesseurs et extracteurs d'attribut sont nécessaires à un système de reconnaissance de caractère performant. Après une description de techniques de préprocesseurs, les différents attibuts toruvés dans la grande quantité de littérature traitant de la reconnaissance de caractère manuscrits sont divisés en deux catégories principales: (1) analyse global et (2) analyse de structure. Une subdivision de ces catégories donne lieu à six familles de types d'attribut, à savoir: (a) distribution de points, (b) transformation, (c) mesures physiques, (d) segments de droite et contours, (e) profil du caractère et (f) ligne centrale du caractère. Chaque famille est décrite à l'aide d'exemples illustratifs. Les performances et les taux de reconnaissance des systèmes utilisant ces attributs sont discutés.</abstract><note type="content">Section title: Regular paper</note><subject><genre>Keywords</genre><topic>Feature extraction</topic><topic>handprint recognition</topic><topic>recognition algorithms</topic></subject><relatedItem type="host"><titleInfo><title>Signal Processing</title></titleInfo><titleInfo type="abbreviated"><title>SIGPRO</title></titleInfo><genre type="Journal">journal</genre><originInfo><dateIssued encoding="w3cdtf">198204</dateIssued></originInfo><identifier type="ISSN">0165-1684</identifier><identifier type="PII">S0165-1684(00)X0293-9</identifier><part><date>198204</date><detail type="volume"><number>4</number><caption>vol.</caption></detail><detail type="issue"><number>2–3</number><caption>no.</caption></detail><extent unit="issue pages"><start>103</start><end>246</end></extent><extent unit="pages"><start>193</start><end>207</end></extent></part></relatedItem><identifier type="istex">72E9014115E3F1E2F0F9B12B3FA520D242150C90</identifier><identifier type="DOI">10.1016/0165-1684(82)90021-4</identifier><identifier type="PII">0165-1684(82)90021-4</identifier><recordInfo><recordContentSource>ELSEVIER</recordContentSource></recordInfo></mods></metadata><enrichments><istex:catWosTEI uri="https://api.istex.fr/document/72E9014115E3F1E2F0F9B12B3FA520D242150C90/enrichments/catWos"><teiHeader><profileDesc><textClass><classCode scheme="WOS">ENGINEERING, ELECTRICAL & ELECTRONIC</classCode></textClass></profileDesc></teiHeader></istex:catWosTEI></enrichments></istex></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Istex/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000333 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Istex/Corpus/biblio.hfd -nk 000333 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Ticri/CIDE
|area= OcrV1
|flux= Istex
|étape= Corpus
|type= RBID
|clé= ISTEX:72E9014115E3F1E2F0F9B12B3FA520D242150C90
|texte= Distinctive features in automatic recognition of handprinted characters
}}
| This area was generated with Dilib version V0.6.32. |  |