Increasing character recognition accuracy by detection and correction of erroneously identified characters
Identifieur interne : 001303 ( Istex/Corpus ); précédent : 001302; suivant : 001304Increasing character recognition accuracy by detection and correction of erroneously identified characters
Auteurs : Cheng Huang Tung ; Hsi-Jian LeeSource :
- Pattern Recognition [ 0031-3203 ] ; 1994.
Abstract
This paper presents a general approach to increasing character recognition accuracy by detection and correction of erroneously identified characters. A high-accuracy two-stage recognition system for recognizing handwritten Chinese text with 5401 categories is demonstrated. In the first stage of the system, two matching modules are applied to recognize an input character simultaneously. The first matching module uses a direction feature and the second uses a generalized feature. A character is rejected at the first stage when the matching results of the two modules are not the same. Because our first stage recognizes most of the input characters correctly and outputs a small number of candidates for each rejected character, a bigram Markov language model in the second stage can choose a candidate with high recognition rate for each rejected character according to contextual information. Experiments are performed on sentences consisting of characters extracted from the CCL/HCCR1 handwritten Chinese character database. In the first stage, the rejection rate for the input characters is 13% and the recognition rate for the accepted characters is 95.9%. In the second stage, the recognition rate for the characters rejected by the first stage is 91.2%. Thus, the overall recognition rate for the input handwritten text is 95.9% × 0.87 + 91.2% × 0.13 (= 95.2%)
Url:
DOI: 10.1016/0031-3203(94)90009-4
Links to Exploration step
ISTEX:119077C352A39B981CC1DED9DF3B2A1BD0226255Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title>Increasing character recognition accuracy by detection and correction of erroneously identified characters</title>
<author><name sortKey="Tung, Cheng Huang" sort="Tung, Cheng Huang" uniqKey="Tung C" first="Cheng Huang" last="Tung">Cheng Huang Tung</name>
<affiliation><mods:affiliation>Department of Computer Science and Information Engineering, National Chiao Tung University, Hsinchu, 30050 Taiwan, Republic of China</mods:affiliation>
</affiliation>
</author>
<author><name sortKey="Lee, Hsi Jian" sort="Lee, Hsi Jian" uniqKey="Lee H" first="Hsi-Jian" last="Lee">Hsi-Jian Lee</name>
<affiliation><mods:affiliation>Department of Computer Science and Information Engineering, National Chiao Tung University, Hsinchu, 30050 Taiwan, Republic of China</mods:affiliation>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">ISTEX</idno>
<idno type="RBID">ISTEX:119077C352A39B981CC1DED9DF3B2A1BD0226255</idno>
<date when="1994" year="1994">1994</date>
<idno type="doi">10.1016/0031-3203(94)90009-4</idno>
<idno type="url">https://api.istex.fr/document/119077C352A39B981CC1DED9DF3B2A1BD0226255/fulltext/pdf</idno>
<idno type="wicri:Area/Istex/Corpus">001303</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title level="a">Increasing character recognition accuracy by detection and correction of erroneously identified characters</title>
<author><name sortKey="Tung, Cheng Huang" sort="Tung, Cheng Huang" uniqKey="Tung C" first="Cheng Huang" last="Tung">Cheng Huang Tung</name>
<affiliation><mods:affiliation>Department of Computer Science and Information Engineering, National Chiao Tung University, Hsinchu, 30050 Taiwan, Republic of China</mods:affiliation>
</affiliation>
</author>
<author><name sortKey="Lee, Hsi Jian" sort="Lee, Hsi Jian" uniqKey="Lee H" first="Hsi-Jian" last="Lee">Hsi-Jian Lee</name>
<affiliation><mods:affiliation>Department of Computer Science and Information Engineering, National Chiao Tung University, Hsinchu, 30050 Taiwan, Republic of China</mods:affiliation>
</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="1994">1994</date>
<biblScope unit="volume">27</biblScope>
<biblScope unit="issue">9</biblScope>
<biblScope unit="page" from="1259">1259</biblScope>
<biblScope unit="page" to="1266">1266</biblScope>
</imprint>
<idno type="ISSN">0031-3203</idno>
</series>
<idno type="istex">119077C352A39B981CC1DED9DF3B2A1BD0226255</idno>
<idno type="DOI">10.1016/0031-3203(94)90009-4</idno>
<idno type="PII">0031-3203(94)90009-4</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">This paper presents a general approach to increasing character recognition accuracy by detection and correction of erroneously identified characters. A high-accuracy two-stage recognition system for recognizing handwritten Chinese text with 5401 categories is demonstrated. In the first stage of the system, two matching modules are applied to recognize an input character simultaneously. The first matching module uses a direction feature and the second uses a generalized feature. A character is rejected at the first stage when the matching results of the two modules are not the same. Because our first stage recognizes most of the input characters correctly and outputs a small number of candidates for each rejected character, a bigram Markov language model in the second stage can choose a candidate with high recognition rate for each rejected character according to contextual information. Experiments are performed on sentences consisting of characters extracted from the CCL/HCCR1 handwritten Chinese character database. In the first stage, the rejection rate for the input characters is 13% and the recognition rate for the accepted characters is 95.9%. In the second stage, the recognition rate for the characters rejected by the first stage is 91.2%. Thus, the overall recognition rate for the input handwritten text is 95.9% × 0.87 + 91.2% × 0.13 (= 95.2%)</div>
</front>
</TEI>
<istex><corpusName>elsevier</corpusName>
<author><json:item><name>Cheng Huang Tung</name>
<affiliations><json:string>Department of Computer Science and Information Engineering, National Chiao Tung University, Hsinchu, 30050 Taiwan, Republic of China</json:string>
</affiliations>
</json:item>
<json:item><name>Hsi-Jian Lee</name>
<affiliations><json:string>Department of Computer Science and Information Engineering, National Chiao Tung University, Hsinchu, 30050 Taiwan, Republic of China</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>Handwritten Chinese characters</value>
</json:item>
<json:item><lang><json:string>eng</json:string>
</lang>
<value>Generalized feature</value>
</json:item>
<json:item><lang><json:string>eng</json:string>
</lang>
<value>Hill-climbing method</value>
</json:item>
<json:item><lang><json:string>eng</json:string>
</lang>
<value>Markov language model</value>
</json:item>
<json:item><lang><json:string>eng</json:string>
</lang>
<value>Part-of-speech</value>
</json:item>
</subject>
<language><json:string>eng</json:string>
</language>
<abstract>This paper presents a general approach to increasing character recognition accuracy by detection and correction of erroneously identified characters. A high-accuracy two-stage recognition system for recognizing handwritten Chinese text with 5401 categories is demonstrated. In the first stage of the system, two matching modules are applied to recognize an input character simultaneously. The first matching module uses a direction feature and the second uses a generalized feature. A character is rejected at the first stage when the matching results of the two modules are not the same. Because our first stage recognizes most of the input characters correctly and outputs a small number of candidates for each rejected character, a bigram Markov language model in the second stage can choose a candidate with high recognition rate for each rejected character according to contextual information. Experiments are performed on sentences consisting of characters extracted from the CCL/HCCR1 handwritten Chinese character database. In the first stage, the rejection rate for the input characters is 13% and the recognition rate for the accepted characters is 95.9%. In the second stage, the recognition rate for the characters rejected by the first stage is 91.2%. Thus, the overall recognition rate for the input handwritten text is 95.9% × 0.87 + 91.2% × 0.13 (= 95.2%)</abstract>
<qualityIndicators><score>6.252</score>
<pdfVersion>1.2</pdfVersion>
<pdfPageSize>540 x 792 pts</pdfPageSize>
<refBibsNative>true</refBibsNative>
<keywordCount>6</keywordCount>
<abstractCharCount>1371</abstractCharCount>
<pdfWordCount>3732</pdfWordCount>
<pdfCharCount>21594</pdfCharCount>
<pdfPageCount>8</pdfPageCount>
<abstractWordCount>210</abstractWordCount>
</qualityIndicators>
<title>Increasing character recognition accuracy by detection and correction of erroneously identified characters</title>
<pii><json:string>0031-3203(94)90009-4</json:string>
</pii>
<genre><json:string>research-article</json:string>
</genre>
<serie><pages><last>167</last>
<first>163</first>
</pages>
<genre></genre>
<language><json:string>unknown</json:string>
</language>
<title>Proc. ICPR</title>
</serie>
<host><volume>27</volume>
<pii><json:string>S0031-3203(00)X0102-1</json:string>
</pii>
<pages><last>1266</last>
<first>1259</first>
</pages>
<issn><json:string>0031-3203</json:string>
</issn>
<issue>9</issue>
<genre><json:string>Journal</json:string>
</genre>
<language><json:string>unknown</json:string>
</language>
<title>Pattern Recognition</title>
<publicationDate>1994</publicationDate>
</host>
<categories><wos><json:string>COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE</json:string>
<json:string>ENGINEERING, ELECTRICAL & ELECTRONIC</json:string>
</wos>
</categories>
<publicationDate>1994</publicationDate>
<copyrightDate>1994</copyrightDate>
<doi><json:string>10.1016/0031-3203(94)90009-4</json:string>
</doi>
<id>119077C352A39B981CC1DED9DF3B2A1BD0226255</id>
<fulltext><json:item><original>true</original>
<mimetype>application/pdf</mimetype>
<extension>pdf</extension>
<uri>https://api.istex.fr/document/119077C352A39B981CC1DED9DF3B2A1BD0226255/fulltext/pdf</uri>
</json:item>
<json:item><original>true</original>
<mimetype>text/plain</mimetype>
<extension>txt</extension>
<uri>https://api.istex.fr/document/119077C352A39B981CC1DED9DF3B2A1BD0226255/fulltext/txt</uri>
</json:item>
<json:item><original>false</original>
<mimetype>application/zip</mimetype>
<extension>zip</extension>
<uri>https://api.istex.fr/document/119077C352A39B981CC1DED9DF3B2A1BD0226255/fulltext/zip</uri>
</json:item>
<istex:fulltextTEI uri="https://api.istex.fr/document/119077C352A39B981CC1DED9DF3B2A1BD0226255/fulltext/tei"><teiHeader><fileDesc><titleStmt><title level="a">Increasing character recognition accuracy by detection and correction of erroneously identified characters</title>
</titleStmt>
<publicationStmt><authority>ISTEX</authority>
<publisher>ELSEVIER</publisher>
<availability><p>ELSEVIER</p>
</availability>
<date>1994</date>
</publicationStmt>
<sourceDesc><biblStruct type="inbook"><analytic><title level="a">Increasing character recognition accuracy by detection and correction of erroneously identified characters</title>
<author><persName><forename type="first">Cheng Huang</forename>
<surname>Tung</surname>
</persName>
<affiliation>Department of Computer Science and Information Engineering, National Chiao Tung University, Hsinchu, 30050 Taiwan, Republic of China</affiliation>
</author>
<author><persName><forename type="first">Hsi-Jian</forename>
<surname>Lee</surname>
</persName>
<affiliation>Author to whom all correspondence should be addressed.</affiliation>
<affiliation>Department of Computer Science and Information Engineering, National Chiao Tung University, Hsinchu, 30050 Taiwan, Republic of China</affiliation>
</author>
</analytic>
<monogr><title level="j">Pattern Recognition</title>
<title level="j" type="abbrev">PR</title>
<idno type="pISSN">0031-3203</idno>
<idno type="PII">S0031-3203(00)X0102-1</idno>
<imprint><publisher>ELSEVIER</publisher>
<date type="published" when="1994"></date>
<biblScope unit="volume">27</biblScope>
<biblScope unit="issue">9</biblScope>
<biblScope unit="page" from="1259">1259</biblScope>
<biblScope unit="page" to="1266">1266</biblScope>
</imprint>
</monogr>
<idno type="istex">119077C352A39B981CC1DED9DF3B2A1BD0226255</idno>
<idno type="DOI">10.1016/0031-3203(94)90009-4</idno>
<idno type="PII">0031-3203(94)90009-4</idno>
</biblStruct>
</sourceDesc>
</fileDesc>
<profileDesc><creation><date>1994</date>
</creation>
<langUsage><language ident="en">en</language>
</langUsage>
<abstract xml:lang="en"><p>This paper presents a general approach to increasing character recognition accuracy by detection and correction of erroneously identified characters. A high-accuracy two-stage recognition system for recognizing handwritten Chinese text with 5401 categories is demonstrated. In the first stage of the system, two matching modules are applied to recognize an input character simultaneously. The first matching module uses a direction feature and the second uses a generalized feature. A character is rejected at the first stage when the matching results of the two modules are not the same. Because our first stage recognizes most of the input characters correctly and outputs a small number of candidates for each rejected character, a bigram Markov language model in the second stage can choose a candidate with high recognition rate for each rejected character according to contextual information. Experiments are performed on sentences consisting of characters extracted from the CCL/HCCR1 handwritten Chinese character database. In the first stage, the rejection rate for the input characters is 13% and the recognition rate for the accepted characters is 95.9%. In the second stage, the recognition rate for the characters rejected by the first stage is 91.2%. Thus, the overall recognition rate for the input handwritten text is 95.9% × 0.87 + 91.2% × 0.13 (= 95.2%)</p>
</abstract>
<textClass><keywords scheme="keyword"><list><head>Keywords</head>
<item><term>Character recognition</term>
</item>
<item><term>Handwritten Chinese characters</term>
</item>
<item><term>Generalized feature</term>
</item>
<item><term>Hill-climbing method</term>
</item>
<item><term>Markov language model</term>
</item>
<item><term>Part-of-speech</term>
</item>
</list>
</keywords>
</textClass>
</profileDesc>
<revisionDesc><change when="1994-04-05">Registration</change>
<change when="1994-03-03">Modified</change>
<change when="1994">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>PR</jid>
<aid>94900094</aid>
<ce:pii>0031-3203(94)90009-4</ce:pii>
<ce:doi>10.1016/0031-3203(94)90009-4</ce:doi>
<ce:copyright type="unknown" year="1994"></ce:copyright>
</item-info>
<head><ce:title>Increasing character recognition accuracy by detection and correction of erroneously identified characters</ce:title>
<ce:author-group><ce:author><ce:given-name>Cheng Huang</ce:given-name>
<ce:surname>Tung</ce:surname>
</ce:author>
<ce:author><ce:given-name>Hsi-Jian</ce:given-name>
<ce:surname>Lee</ce:surname>
<ce:cross-ref refid="COR1"><ce:sup>†</ce:sup>
</ce:cross-ref>
</ce:author>
<ce:affiliation><ce:textfn>Department of Computer Science and Information Engineering, National Chiao Tung University, Hsinchu, 30050 Taiwan, Republic of China</ce:textfn>
</ce:affiliation>
<ce:correspondence id="COR1"><ce:label>†</ce:label>
<ce:text>Author to whom all correspondence should be addressed.</ce:text>
</ce:correspondence>
</ce:author-group>
<ce:date-received day="1" month="9" year="1993"></ce:date-received>
<ce:date-revised day="3" month="3" year="1994"></ce:date-revised>
<ce:date-accepted day="5" month="4" year="1994"></ce:date-accepted>
<ce:abstract><ce:section-title>Abstract</ce:section-title>
<ce:abstract-sec><ce:simple-para>This paper presents a general approach to increasing character recognition accuracy by detection and correction of erroneously identified characters. A high-accuracy two-stage recognition system for recognizing handwritten Chinese text with 5401 categories is demonstrated. In the first stage of the system, two matching modules are applied to recognize an input character simultaneously. The first matching module uses a direction feature and the second uses a generalized feature. A character is rejected at the first stage when the matching results of the two modules are not the same. Because our first stage recognizes most of the input characters correctly and outputs a small number of candidates for each rejected character, a bigram Markov language model in the second stage can choose a candidate with high recognition rate for each rejected character according to contextual information. Experiments are performed on sentences consisting of characters extracted from the CCL/HCCR1 handwritten Chinese character database. In the first stage, the rejection rate for the input characters is 13% and the recognition rate for the accepted characters is 95.9%. In the second stage, the recognition rate for the characters rejected by the first stage is 91.2%. Thus, the overall recognition rate for the input handwritten text is 95.9% × 0.87 + 91.2% × 0.13 (= 95.2%)</ce:simple-para>
</ce:abstract-sec>
</ce:abstract>
<ce:keywords><ce:section-title>Keywords</ce:section-title>
<ce:keyword><ce:text>Character recognition</ce:text>
</ce:keyword>
<ce:keyword><ce:text>Handwritten Chinese characters</ce:text>
</ce:keyword>
<ce:keyword><ce:text>Generalized feature</ce:text>
</ce:keyword>
<ce:keyword><ce:text>Hill-climbing method</ce:text>
</ce:keyword>
<ce:keyword><ce:text>Markov language model</ce:text>
</ce:keyword>
<ce:keyword><ce:text>Part-of-speech</ce:text>
</ce:keyword>
</ce:keywords>
</head>
</converted-article>
</istex:document>
</istex:metadataXml>
<mods version="3.6"><titleInfo><title>Increasing character recognition accuracy by detection and correction of erroneously identified characters</title>
</titleInfo>
<titleInfo type="alternative" contentType="CDATA"><title>Increasing character recognition accuracy by detection and correction of erroneously identified characters</title>
</titleInfo>
<name type="personal"><namePart type="given">Cheng Huang</namePart>
<namePart type="family">Tung</namePart>
<affiliation>Department of Computer Science and Information Engineering, National Chiao Tung University, Hsinchu, 30050 Taiwan, Republic of China</affiliation>
<role><roleTerm type="text">author</roleTerm>
</role>
</name>
<name type="personal"><namePart type="given">Hsi-Jian</namePart>
<namePart type="family">Lee</namePart>
<affiliation>Department of Computer Science and Information Engineering, National Chiao Tung University, Hsinchu, 30050 Taiwan, Republic of China</affiliation>
<description>Author to whom all correspondence should be addressed.</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">1994</dateIssued>
<dateValid encoding="w3cdtf">1994-04-05</dateValid>
<dateModified encoding="w3cdtf">1994-03-03</dateModified>
<copyrightDate encoding="w3cdtf">1994</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">This paper presents a general approach to increasing character recognition accuracy by detection and correction of erroneously identified characters. A high-accuracy two-stage recognition system for recognizing handwritten Chinese text with 5401 categories is demonstrated. In the first stage of the system, two matching modules are applied to recognize an input character simultaneously. The first matching module uses a direction feature and the second uses a generalized feature. A character is rejected at the first stage when the matching results of the two modules are not the same. Because our first stage recognizes most of the input characters correctly and outputs a small number of candidates for each rejected character, a bigram Markov language model in the second stage can choose a candidate with high recognition rate for each rejected character according to contextual information. Experiments are performed on sentences consisting of characters extracted from the CCL/HCCR1 handwritten Chinese character database. In the first stage, the rejection rate for the input characters is 13% and the recognition rate for the accepted characters is 95.9%. In the second stage, the recognition rate for the characters rejected by the first stage is 91.2%. Thus, the overall recognition rate for the input handwritten text is 95.9% × 0.87 + 91.2% × 0.13 (= 95.2%)</abstract>
<subject><genre>Keywords</genre>
<topic>Character recognition</topic>
<topic>Handwritten Chinese characters</topic>
<topic>Generalized feature</topic>
<topic>Hill-climbing method</topic>
<topic>Markov language model</topic>
<topic>Part-of-speech</topic>
</subject>
<relatedItem type="host"><titleInfo><title>Pattern Recognition</title>
</titleInfo>
<titleInfo type="abbreviated"><title>PR</title>
</titleInfo>
<genre type="Journal">journal</genre>
<originInfo><dateIssued encoding="w3cdtf">199409</dateIssued>
</originInfo>
<identifier type="ISSN">0031-3203</identifier>
<identifier type="PII">S0031-3203(00)X0102-1</identifier>
<part><date>199409</date>
<detail type="volume"><number>27</number>
<caption>vol.</caption>
</detail>
<detail type="issue"><number>9</number>
<caption>no.</caption>
</detail>
<extent unit="issue pages"><start>1127</start>
<end>1289</end>
</extent>
<extent unit="pages"><start>1259</start>
<end>1266</end>
</extent>
</part>
</relatedItem>
<identifier type="istex">119077C352A39B981CC1DED9DF3B2A1BD0226255</identifier>
<identifier type="DOI">10.1016/0031-3203(94)90009-4</identifier>
<identifier type="PII">0031-3203(94)90009-4</identifier>
<recordInfo><recordContentSource>ELSEVIER</recordContentSource>
</recordInfo>
</mods>
</metadata>
<enrichments><istex:catWosTEI uri="https://api.istex.fr/document/119077C352A39B981CC1DED9DF3B2A1BD0226255/enrichments/catWos"><teiHeader><profileDesc><textClass><classCode scheme="WOS">COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE</classCode>
<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 001303 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Istex/Corpus/biblio.hfd -nk 001303 | 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:119077C352A39B981CC1DED9DF3B2A1BD0226255 |texte= Increasing character recognition accuracy by detection and correction of erroneously identified characters }}
![]() | This area was generated with Dilib version V0.6.32. | ![]() |