Machine-printed character recognition revisited: re-application of recent advances in handwritten character recognition research
Identifieur interne : 001497 ( Istex/Corpus ); précédent : 001496; suivant : 001498Machine-printed character recognition revisited: re-application of recent advances in handwritten character recognition research
Auteurs : A. F. R. Rahman ; M. C. FairhurstSource :
- Image and Vision Computing [ 0262-8856 ] ; 1997.
Abstract
The increasing automation of document processing applications has emphasised the need for robust and reliable recognition of machine-printed characters. Although research in character recognition is now focused principally on applications in reading handwritten characters, this paper demonstrates how recent progress in the area of multiple-expert classification can be exploited to provide new approaches to the processing of printed data. Established classification algorithms have been used in a new multiple-expert framework to generate an optimised decision combination platform comprising multiple experts, and significant improvements in overall recognition performance on machine-printed characters have been achieved.
Url:
DOI: 10.1016/S0262-8856(98)00056-0
Links to Exploration step
ISTEX:78AA2A344578D4547448310260102FF1F7E3208ELe document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title>Machine-printed character recognition revisited: re-application of recent advances in handwritten character recognition research</title><author><name sortKey="Rahman, A F R" sort="Rahman, A F R" uniqKey="Rahman A" first="A. F. R." last="Rahman">A. F. R. Rahman</name><affiliation><mods:affiliation>Electronic Engineering Laboratories, University of Kent, Canterbury, Kent CT2 7NT, UK</mods:affiliation></affiliation></author><author><name sortKey="Fairhurst, M C" sort="Fairhurst, M C" uniqKey="Fairhurst M" first="M. C." last="Fairhurst">M. C. Fairhurst</name><affiliation><mods:affiliation>Electronic Engineering Laboratories, University of Kent, Canterbury, Kent CT2 7NT, UK</mods:affiliation></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:78AA2A344578D4547448310260102FF1F7E3208E</idno><date when="1998" year="1998">1998</date><idno type="doi">10.1016/S0262-8856(98)00056-0</idno><idno type="url">https://api.istex.fr/document/78AA2A344578D4547448310260102FF1F7E3208E/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">001497</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a">Machine-printed character recognition revisited: re-application of recent advances in handwritten character recognition research</title><author><name sortKey="Rahman, A F R" sort="Rahman, A F R" uniqKey="Rahman A" first="A. F. R." last="Rahman">A. F. R. Rahman</name><affiliation><mods:affiliation>Electronic Engineering Laboratories, University of Kent, Canterbury, Kent CT2 7NT, UK</mods:affiliation></affiliation></author><author><name sortKey="Fairhurst, M C" sort="Fairhurst, M C" uniqKey="Fairhurst M" first="M. C." last="Fairhurst">M. C. Fairhurst</name><affiliation><mods:affiliation>Electronic Engineering Laboratories, University of Kent, Canterbury, Kent CT2 7NT, UK</mods:affiliation></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="1997">1997</date><biblScope unit="volume">16</biblScope><biblScope unit="issue">12–13</biblScope><biblScope unit="page" from="819">819</biblScope><biblScope unit="page" to="842">842</biblScope></imprint><idno type="ISSN">0262-8856</idno></series><idno type="istex">78AA2A344578D4547448310260102FF1F7E3208E</idno><idno type="DOI">10.1016/S0262-8856(98)00056-0</idno><idno type="PII">S0262-8856(98)00056-0</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">The increasing automation of document processing applications has emphasised the need for robust and reliable recognition of machine-printed characters. Although research in character recognition is now focused principally on applications in reading handwritten characters, this paper demonstrates how recent progress in the area of multiple-expert classification can be exploited to provide new approaches to the processing of printed data. Established classification algorithms have been used in a new multiple-expert framework to generate an optimised decision combination platform comprising multiple experts, and significant improvements in overall recognition performance on machine-printed characters have been achieved.</div></front></TEI><istex><corpusName>elsevier</corpusName><author><json:item><name>A.F.R. Rahman</name><affiliations><json:string>Electronic Engineering Laboratories, University of Kent, Canterbury, Kent CT2 7NT, UK</json:string></affiliations></json:item><json:item><name>M.C. Fairhurst</name><affiliations><json:string>Electronic Engineering Laboratories, University of Kent, Canterbury, Kent CT2 7NT, UK</json:string></affiliations></json:item></author><subject><json:item><lang><json:string>eng</json:string></lang><value>Character recognition</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Multiple-expert classifiers</value></json:item><json:item><lang><json:string>eng</json:string></lang><value>Combined decision platform</value></json:item></subject><language><json:string>eng</json:string></language><abstract>The increasing automation of document processing applications has emphasised the need for robust and reliable recognition of machine-printed characters. Although research in character recognition is now focused principally on applications in reading handwritten characters, this paper demonstrates how recent progress in the area of multiple-expert classification can be exploited to provide new approaches to the processing of printed data. Established classification algorithms have been used in a new multiple-expert framework to generate an optimised decision combination platform comprising multiple experts, and significant improvements in overall recognition performance on machine-printed characters have been achieved.</abstract><qualityIndicators><score>6.116</score><pdfVersion>1.2</pdfVersion><pdfPageSize>596 x 795 pts</pdfPageSize><refBibsNative>true</refBibsNative><keywordCount>3</keywordCount><abstractCharCount>727</abstractCharCount><pdfWordCount>13683</pdfWordCount><pdfCharCount>87647</pdfCharCount><pdfPageCount>24</pdfPageCount><abstractWordCount>93</abstractWordCount></qualityIndicators><title>Machine-printed character recognition revisited: re-application of recent advances in handwritten character recognition research</title><pii><json:string>S0262-8856(98)00056-0</json:string></pii><genre><json:string>research-article</json:string></genre><host><volume>16</volume><pii><json:string>S0262-8856(00)X0041-8</json:string></pii><pages><last>842</last><first>819</first></pages><issn><json:string>0262-8856</json:string></issn><issue>12–13</issue><genre><json:string>Journal</json:string></genre><language><json:string>unknown</json:string></language><title>Image and Vision Computing</title><publicationDate>1998</publicationDate></host><categories><wos><json:string>COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE</json:string><json:string>COMPUTER SCIENCE, SOFTWARE ENGINEERING</json:string><json:string>COMPUTER SCIENCE, THEORY & METHODS</json:string><json:string>ENGINEERING, ELECTRICAL & ELECTRONIC</json:string><json:string>OPTICS</json:string></wos></categories><publicationDate>1997</publicationDate><copyrightDate>1998</copyrightDate><doi><json:string>10.1016/S0262-8856(98)00056-0</json:string></doi><id>78AA2A344578D4547448310260102FF1F7E3208E</id><fulltext><json:item><original>true</original><mimetype>application/pdf</mimetype><extension>pdf</extension><uri>https://api.istex.fr/document/78AA2A344578D4547448310260102FF1F7E3208E/fulltext/pdf</uri></json:item><json:item><original>true</original><mimetype>text/plain</mimetype><extension>txt</extension><uri>https://api.istex.fr/document/78AA2A344578D4547448310260102FF1F7E3208E/fulltext/txt</uri></json:item><json:item><original>false</original><mimetype>application/zip</mimetype><extension>zip</extension><uri>https://api.istex.fr/document/78AA2A344578D4547448310260102FF1F7E3208E/fulltext/zip</uri></json:item><istex:fulltextTEI uri="https://api.istex.fr/document/78AA2A344578D4547448310260102FF1F7E3208E/fulltext/tei"><teiHeader><fileDesc><titleStmt><title level="a">Machine-printed character recognition revisited: re-application of recent advances in handwritten character recognition research</title></titleStmt><publicationStmt><authority>ISTEX</authority><publisher>ELSEVIER</publisher><availability><p>ELSEVIER</p></availability><date>1998</date></publicationStmt><notesStmt><note type="content">Fig. 1: Single perceptron.</note><note type="content">Fig. 2: Multilayer perceptron network.</note><note type="content">Fig. 3: Parallel combination of multiple experts.</note><note type="content">Fig. 4: Decision combination of multiple experts.</note><note type="content">Fig. 5: Serial combination of multiple experts.</note><note type="content">Fig. 6: The structure of the multiple-expert classifier.</note><note type="content">Fig. 7: Some typical examples from the reference machine-printed database.</note><note type="content">Fig. 8: Some typical examples from the reference hand-printed database.</note><note type="content">Fig. 9: Comparative performance enhancement on different datasets: the different configurations have been ranked between 0 and 13 for illustrative purposes.</note><note type="content">Fig. 10: Throughput: comparison of different configurations.</note><note type="content">Fig. 11: Profile of processing time vs. percentage recognition rate on the printed database: (a) on digit classes; (b) on digit plus upper case classes.</note><note type="content">Table 1: Performance of the single classifiers, BWS, FWS and MPC</note><note type="content">Table 2: Performance of the single classifier, MLP</note><note type="content">Table 3: Performance of the Type-I expert combination approach</note><note type="content">Table 4: Performance of the Type-II expert combination approach</note><note type="content">Table 5: Details of the Type-II expert combination approach</note><note type="content">Table 6: Performance of the Type-III expert combination approach</note><note type="content">Table 7: Performance of the block of filters</note><note type="content">Table 8: Performance of the specialised classifiers (dichotomisers/trichotomisers)</note><note type="content">Table 9: Enhancement comparison with respect to single-expert classifiers</note><note type="content">Table 10: Performance of individual experts on the handwritten characters</note><note type="content">Table 11: Enhancement comparison with respect to the handwritten database</note><note type="content">Table 12: Enhancement comparison with respect to the handwritten database</note><note type="content">Table 13: Ranking of the different configurations</note><note type="content">Table 14: Enhancement comparison with respect to the handwritten database</note><note type="content">Table 15: Comparison among different configurations with respect to the processing time</note></notesStmt><sourceDesc><biblStruct type="inbook"><analytic><title level="a">Machine-printed character recognition revisited: re-application of recent advances in handwritten character recognition research</title><author><persName><forename type="first">A.F.R.</forename><surname>Rahman</surname></persName><affiliation>Electronic Engineering Laboratories, University of Kent, Canterbury, Kent CT2 7NT, UK</affiliation></author><author><persName><forename type="first">M.C.</forename><surname>Fairhurst</surname></persName><note type="correspondence"><p>Corresponding author.</p></note><affiliation>Electronic Engineering Laboratories, University of Kent, Canterbury, Kent CT2 7NT, UK</affiliation></author></analytic><monogr><title level="j">Image and Vision Computing</title><title level="j" type="abbrev">IMAVIS</title><idno type="pISSN">0262-8856</idno><idno type="PII">S0262-8856(00)X0041-8</idno><imprint><publisher>ELSEVIER</publisher><date type="published" when="1997"></date><biblScope unit="volume">16</biblScope><biblScope unit="issue">12–13</biblScope><biblScope unit="page" from="819">819</biblScope><biblScope unit="page" to="842">842</biblScope></imprint></monogr><idno type="istex">78AA2A344578D4547448310260102FF1F7E3208E</idno><idno type="DOI">10.1016/S0262-8856(98)00056-0</idno><idno type="PII">S0262-8856(98)00056-0</idno></biblStruct></sourceDesc></fileDesc><profileDesc><creation><date>1998</date></creation><langUsage><language ident="en">en</language></langUsage><abstract xml:lang="en"><p>The increasing automation of document processing applications has emphasised the need for robust and reliable recognition of machine-printed characters. Although research in character recognition is now focused principally on applications in reading handwritten characters, this paper demonstrates how recent progress in the area of multiple-expert classification can be exploited to provide new approaches to the processing of printed data. Established classification algorithms have been used in a new multiple-expert framework to generate an optimised decision combination platform comprising multiple experts, and significant improvements in overall recognition performance on machine-printed characters have been achieved.</p></abstract><textClass><keywords scheme="keyword"><list><head>Keywords</head><item><term>Character recognition</term></item><item><term>Multiple-expert classifiers</term></item><item><term>Combined decision platform</term></item></list></keywords></textClass></profileDesc><revisionDesc><change when="1997-11-12">Registration</change><change when="1997-09-23">Modified</change><change when="1997">Published</change></revisionDesc></teiHeader></istex:fulltextTEI></fulltext><metadata><istex:metadataXml wicri:clean="Elsevier, elements deleted: ce:floats; body; 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:entity SYSTEM="gr1" NDATA="IMAGE" name="gr1"></istex:entity><istex:entity SYSTEM="gr2" NDATA="IMAGE" name="gr2"></istex:entity><istex:entity SYSTEM="gr3" NDATA="IMAGE" name="gr3"></istex:entity><istex:entity SYSTEM="gr4" NDATA="IMAGE" name="gr4"></istex:entity><istex:entity SYSTEM="gr5" NDATA="IMAGE" name="gr5"></istex:entity><istex:entity SYSTEM="gr6" NDATA="IMAGE" name="gr6"></istex:entity><istex:entity SYSTEM="gr7" NDATA="IMAGE" name="gr7"></istex:entity><istex:entity SYSTEM="gr8" NDATA="IMAGE" name="gr8"></istex:entity><istex:entity SYSTEM="gr9" NDATA="IMAGE" name="gr9"></istex:entity><istex:entity SYSTEM="gr10" NDATA="IMAGE" name="gr10"></istex:entity><istex:entity SYSTEM="gr11" NDATA="IMAGE" name="gr11"></istex:entity></istex:docType><istex:document><converted-article version="4.5.2" docsubtype="fla"><item-info><jid>IMAVIS</jid><aid>1512</aid><ce:pii>S0262-8856(98)00056-0</ce:pii><ce:doi>10.1016/S0262-8856(98)00056-0</ce:doi><ce:copyright year="1998" type="full-transfer">Elsevier Science B.V.</ce:copyright></item-info><head><ce:title>Machine-printed character recognition revisited: re-application of recent advances in handwritten character recognition research</ce:title><ce:author-group><ce:author><ce:given-name>A.F.R.</ce:given-name><ce:surname>Rahman</ce:surname></ce:author><ce:author><ce:given-name>M.C.</ce:given-name><ce:surname>Fairhurst</ce:surname><ce:cross-ref refid="CORR1">*</ce:cross-ref></ce:author><ce:affiliation id="AFF1"><ce:label>affl1</ce:label><ce:textfn>Electronic Engineering Laboratories, University of Kent, Canterbury, Kent CT2 7NT, UK</ce:textfn></ce:affiliation><ce:correspondence id="CORR1"><ce:label>*</ce:label><ce:text>Corresponding author.</ce:text></ce:correspondence></ce:author-group><ce:date-received day="21" month="3" year="1997"></ce:date-received><ce:date-revised day="23" month="9" year="1997"></ce:date-revised><ce:date-accepted day="12" month="11" year="1997"></ce:date-accepted><ce:abstract><ce:section-title>Abstract</ce:section-title><ce:abstract-sec><ce:simple-para>The increasing automation of document processing applications has emphasised the need for robust and reliable recognition of machine-printed characters. Although research in character recognition is now focused principally on applications in reading handwritten characters, this paper demonstrates how recent progress in the area of multiple-expert classification can be exploited to provide new approaches to the processing of printed data. Established classification algorithms have been used in a new multiple-expert framework to generate an optimised decision combination platform comprising multiple experts, and significant improvements in overall recognition performance on machine-printed characters have been achieved.</ce:simple-para></ce:abstract-sec></ce:abstract><ce:keywords class="keyword"><ce:section-title>Keywords</ce:section-title><ce:keyword><ce:text>Character recognition</ce:text></ce:keyword><ce:keyword><ce:text>Multiple-expert classifiers</ce:text></ce:keyword><ce:keyword><ce:text>Combined decision platform</ce:text></ce:keyword></ce:keywords></head></converted-article></istex:document></istex:metadataXml><mods version="3.6"><titleInfo><title>Machine-printed character recognition revisited: re-application of recent advances in handwritten character recognition research</title></titleInfo><titleInfo type="alternative" contentType="CDATA"><title>Machine-printed character recognition revisited: re-application of recent advances in handwritten character recognition research</title></titleInfo><name type="personal"><namePart type="given">A.F.R.</namePart><namePart type="family">Rahman</namePart><affiliation>Electronic Engineering Laboratories, University of Kent, Canterbury, Kent CT2 7NT, UK</affiliation><role><roleTerm type="text">author</roleTerm></role></name><name type="personal"><namePart type="given">M.C.</namePart><namePart type="family">Fairhurst</namePart><affiliation>Electronic Engineering Laboratories, University of Kent, Canterbury, Kent CT2 7NT, UK</affiliation><description>Corresponding author.</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">1997</dateIssued><dateValid encoding="w3cdtf">1997-11-12</dateValid><dateModified encoding="w3cdtf">1997-09-23</dateModified><copyrightDate encoding="w3cdtf">1998</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">The increasing automation of document processing applications has emphasised the need for robust and reliable recognition of machine-printed characters. Although research in character recognition is now focused principally on applications in reading handwritten characters, this paper demonstrates how recent progress in the area of multiple-expert classification can be exploited to provide new approaches to the processing of printed data. Established classification algorithms have been used in a new multiple-expert framework to generate an optimised decision combination platform comprising multiple experts, and significant improvements in overall recognition performance on machine-printed characters have been achieved.</abstract><note type="content">Fig. 1: Single perceptron.</note><note type="content">Fig. 2: Multilayer perceptron network.</note><note type="content">Fig. 3: Parallel combination of multiple experts.</note><note type="content">Fig. 4: Decision combination of multiple experts.</note><note type="content">Fig. 5: Serial combination of multiple experts.</note><note type="content">Fig. 6: The structure of the multiple-expert classifier.</note><note type="content">Fig. 7: Some typical examples from the reference machine-printed database.</note><note type="content">Fig. 8: Some typical examples from the reference hand-printed database.</note><note type="content">Fig. 9: Comparative performance enhancement on different datasets: the different configurations have been ranked between 0 and 13 for illustrative purposes.</note><note type="content">Fig. 10: Throughput: comparison of different configurations.</note><note type="content">Fig. 11: Profile of processing time vs. percentage recognition rate on the printed database: (a) on digit classes; (b) on digit plus upper case classes.</note><note type="content">Table 1: Performance of the single classifiers, BWS, FWS and MPC</note><note type="content">Table 2: Performance of the single classifier, MLP</note><note type="content">Table 3: Performance of the Type-I expert combination approach</note><note type="content">Table 4: Performance of the Type-II expert combination approach</note><note type="content">Table 5: Details of the Type-II expert combination approach</note><note type="content">Table 6: Performance of the Type-III expert combination approach</note><note type="content">Table 7: Performance of the block of filters</note><note type="content">Table 8: Performance of the specialised classifiers (dichotomisers/trichotomisers)</note><note type="content">Table 9: Enhancement comparison with respect to single-expert classifiers</note><note type="content">Table 10: Performance of individual experts on the handwritten characters</note><note type="content">Table 11: Enhancement comparison with respect to the handwritten database</note><note type="content">Table 12: Enhancement comparison with respect to the handwritten database</note><note type="content">Table 13: Ranking of the different configurations</note><note type="content">Table 14: Enhancement comparison with respect to the handwritten database</note><note type="content">Table 15: Comparison among different configurations with respect to the processing time</note><subject><genre>Keywords</genre><topic>Character recognition</topic><topic>Multiple-expert classifiers</topic><topic>Combined decision platform</topic></subject><relatedItem type="host"><titleInfo><title>Image and Vision Computing</title></titleInfo><titleInfo type="abbreviated"><title>IMAVIS</title></titleInfo><genre type="Journal">journal</genre><originInfo><dateIssued encoding="w3cdtf">19980824</dateIssued></originInfo><identifier type="ISSN">0262-8856</identifier><identifier type="PII">S0262-8856(00)X0041-8</identifier><part><date>19980824</date><detail type="volume"><number>16</number><caption>vol.</caption></detail><detail type="issue"><number>12–13</number><caption>no.</caption></detail><extent unit="issue pages"><start>817</start><end>960</end></extent><extent unit="pages"><start>819</start><end>842</end></extent></part></relatedItem><identifier type="istex">78AA2A344578D4547448310260102FF1F7E3208E</identifier><identifier type="DOI">10.1016/S0262-8856(98)00056-0</identifier><identifier type="PII">S0262-8856(98)00056-0</identifier><accessCondition type="use and reproduction" contentType="">© 1998Elsevier Science B.V.</accessCondition><recordInfo><recordContentSource>ELSEVIER</recordContentSource><recordOrigin>Elsevier Science B.V., ©1998</recordOrigin></recordInfo></mods></metadata><enrichments><istex:catWosTEI uri="https://api.istex.fr/document/78AA2A344578D4547448310260102FF1F7E3208E/enrichments/catWos"><teiHeader><profileDesc><textClass><classCode scheme="WOS">COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE</classCode><classCode scheme="WOS">COMPUTER SCIENCE, SOFTWARE ENGINEERING</classCode><classCode scheme="WOS">COMPUTER SCIENCE, THEORY & METHODS</classCode><classCode scheme="WOS">ENGINEERING, ELECTRICAL & ELECTRONIC</classCode><classCode scheme="WOS">OPTICS</classCode></textClass></profileDesc></teiHeader></istex:catWosTEI></enrichments><serie></serie></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 001497 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Istex/Corpus/biblio.hfd -nk 001497 | 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:78AA2A344578D4547448310260102FF1F7E3208E
|texte= Machine-printed character recognition revisited: re-application of recent advances in handwritten character recognition research
}}
| This area was generated with Dilib version V0.6.32. |  |