Text Detection in Chart Images
Identifieur interne : 000183 ( Main/Exploration ); précédent : 000182; suivant : 000184Text Detection in Chart Images
Auteurs : N. Vassilieva [Russie] ; Y. Fomina [Russie]Source :
- Pattern recognition and image analysis [ 1054-6618 ] ; 2013.
Descripteurs français
- Pascal (Inist)
English descriptors
- KwdEn :
Abstract
Common OCR (Optical Character Recognition) systems fail to detect and recognize small text strings of few characters, in particular when a text line is not horizontal. Such text regions are typical for chart images. In this paper we present an algorithm that is able to detect small text regions regardless of string orientation and font size or style. We propose to use this algorithm as a preprocessing step for text recognition with a common OCR engine. According to our experimental results, one can get up to 20 times better text recognition rate, and 15 times higher text recognition precision when the proposed algorithm is used to detect text location, size and orientation, before using an OCR system. Experiments have been performed on a benchmark set of 1000 chart images created with the XML/SWF Chart tool, which contain about 14000 text regions in total.
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream PascalFrancis, to step Corpus: 000032
- to stream PascalFrancis, to step Curation: 000732
- to stream PascalFrancis, to step Checkpoint: 000026
- to stream Main, to step Merge: 000186
- to stream Main, to step Curation: 000183
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">Text Detection in Chart Images</title>
<author><name sortKey="Vassilieva, N" sort="Vassilieva, N" uniqKey="Vassilieva N" first="N." last="Vassilieva">N. Vassilieva</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>HP Labs, I Antillorijskaya str.</s1>
<s2>St. Petersburg, 191104</s2>
<s3>RUS</s3>
<sZ>1 aut.</sZ>
</inist:fA14>
<country>Russie</country>
<wicri:noRegion>St. Petersburg, 191104</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Fomina, Y" sort="Fomina, Y" uniqKey="Fomina Y" first="Y." last="Fomina">Y. Fomina</name>
<affiliation wicri:level="1"><inist:fA14 i1="02"><s1>Studio Mobile, 18A Bolshoy Prospekt</s1>
<s2>St. Petersburg, 197198</s2>
<s3>RUS</s3>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>Russie</country>
<wicri:noRegion>St. Petersburg, 197198</wicri:noRegion>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">INIST</idno>
<idno type="inist">14-0004028</idno>
<date when="2013">2013</date>
<idno type="stanalyst">PASCAL 14-0004028 INIST</idno>
<idno type="RBID">Pascal:14-0004028</idno>
<idno type="wicri:Area/PascalFrancis/Corpus">000032</idno>
<idno type="wicri:Area/PascalFrancis/Curation">000732</idno>
<idno type="wicri:Area/PascalFrancis/Checkpoint">000026</idno>
<idno type="wicri:doubleKey">1054-6618:2013:Vassilieva N:text:detection:in</idno>
<idno type="wicri:Area/Main/Merge">000186</idno>
<idno type="wicri:Area/Main/Curation">000183</idno>
<idno type="wicri:Area/Main/Exploration">000183</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">Text Detection in Chart Images</title>
<author><name sortKey="Vassilieva, N" sort="Vassilieva, N" uniqKey="Vassilieva N" first="N." last="Vassilieva">N. Vassilieva</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>HP Labs, I Antillorijskaya str.</s1>
<s2>St. Petersburg, 191104</s2>
<s3>RUS</s3>
<sZ>1 aut.</sZ>
</inist:fA14>
<country>Russie</country>
<wicri:noRegion>St. Petersburg, 191104</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Fomina, Y" sort="Fomina, Y" uniqKey="Fomina Y" first="Y." last="Fomina">Y. Fomina</name>
<affiliation wicri:level="1"><inist:fA14 i1="02"><s1>Studio Mobile, 18A Bolshoy Prospekt</s1>
<s2>St. Petersburg, 197198</s2>
<s3>RUS</s3>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>Russie</country>
<wicri:noRegion>St. Petersburg, 197198</wicri:noRegion>
</affiliation>
</author>
</analytic>
<series><title level="j" type="main">Pattern recognition and image analysis</title>
<title level="j" type="abbreviated">Pattern recognit. image anal.</title>
<idno type="ISSN">1054-6618</idno>
<imprint><date when="2013">2013</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><title level="j" type="main">Pattern recognition and image analysis</title>
<title level="j" type="abbreviated">Pattern recognit. image anal.</title>
<idno type="ISSN">1054-6618</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Character recognition</term>
<term>Character string</term>
<term>Connected graph</term>
<term>Experimental result</term>
<term>Hough transformation</term>
<term>Information retrieval</term>
<term>Localization</term>
<term>Optical character recognition</term>
<term>Pattern recognition</term>
<term>Text</term>
<term>XML language</term>
</keywords>
<keywords scheme="Pascal" xml:lang="fr"><term>Recherche information</term>
<term>Texte</term>
<term>Reconnaissance optique caractère</term>
<term>Reconnaissance caractère</term>
<term>Chaîne caractère</term>
<term>Reconnaissance forme</term>
<term>Localisation</term>
<term>Langage XML</term>
<term>Résultat expérimental</term>
<term>Graphe connexe</term>
<term>Transformation Hough</term>
<term>.</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">Common OCR (Optical Character Recognition) systems fail to detect and recognize small text strings of few characters, in particular when a text line is not horizontal. Such text regions are typical for chart images. In this paper we present an algorithm that is able to detect small text regions regardless of string orientation and font size or style. We propose to use this algorithm as a preprocessing step for text recognition with a common OCR engine. According to our experimental results, one can get up to 20 times better text recognition rate, and 15 times higher text recognition precision when the proposed algorithm is used to detect text location, size and orientation, before using an OCR system. Experiments have been performed on a benchmark set of 1000 chart images created with the XML/SWF Chart tool, which contain about 14000 text regions in total.</div>
</front>
</TEI>
<affiliations><list><country><li>Russie</li>
</country>
</list>
<tree><country name="Russie"><noRegion><name sortKey="Vassilieva, N" sort="Vassilieva, N" uniqKey="Vassilieva N" first="N." last="Vassilieva">N. Vassilieva</name>
</noRegion>
<name sortKey="Fomina, Y" sort="Fomina, Y" uniqKey="Fomina Y" first="Y." last="Fomina">Y. Fomina</name>
</country>
</tree>
</affiliations>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000183 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000183 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Ticri/CIDE |area= OcrV1 |flux= Main |étape= Exploration |type= RBID |clé= Pascal:14-0004028 |texte= Text Detection in Chart Images }}
This area was generated with Dilib version V0.6.32. |