Mathematical symbol recognition with support vector machines
Identifieur interne : 000D11 ( Main/Exploration ); précédent : 000D10; suivant : 000D12Mathematical symbol recognition with support vector machines
Auteurs : Christopher Malon [Japon] ; Seiichi Uchida [Japon] ; Masakazu Suzuki (mathématicien) [Japon]Source :
- Pattern recognition letters [ 0167-8655 ] ; 2008.
Descripteurs français
- Pascal (Inist)
English descriptors
- KwdEn :
Abstract
Single-character recognition of mathematical symbols poses challenges from its two-dimensional pattern, the variety of similar symbols that must be recognized distinctly, the imbalance and paucity of training data available, and the impossibility of final verification through spell check. We investigate the use of support vector machines to improve the classification of InftyReader, a free system for the OCR of mathematical documents. First, we compare the performance of SVM kernels and feature definitions on pairs of letters that InftyReader usually confuses. Second, we describe a successful approach to multi-class classification with SVM, utilizing the ranking of alternatives within InftyReader's confusion clusters. The inclusion of our technique in InftyReader reduces its misrecognition rate by 41%.
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream PascalFrancis, to step Corpus: 000279
- to stream PascalFrancis, to step Curation: 000505
- to stream PascalFrancis, to step Checkpoint: 000233
- to stream Main, to step Merge: 000D23
- to stream Main, to step Curation: 000D11
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">Mathematical symbol recognition with support vector machines</title>
<author><name sortKey="Malon, Christopher" sort="Malon, Christopher" uniqKey="Malon C" first="Christopher" last="Malon">Christopher Malon</name>
<affiliation wicri:level="4"><inist:fA14 i1="01"><s1>Faculty of Mathematics, Kyushu University, Hakozaki 6-10-1</s1>
<s2>Higashi-ku, Fukuoka 812-8581</s2>
<s3>JPN</s3>
<sZ>1 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Japon</country>
<wicri:noRegion>Higashi-ku, Fukuoka 812-8581</wicri:noRegion>
<orgName type="university">Université de Kyūshū</orgName>
<placeName><settlement type="city">Fukuoka</settlement>
<region type="province">Kyūshū</region>
<region type="prefecture">Préfecture de Fukuoka</region>
</placeName>
</affiliation>
</author>
<author><name sortKey="Uchida, Seiichi" sort="Uchida, Seiichi" uniqKey="Uchida S" first="Seiichi" last="Uchida">Seiichi Uchida</name>
<affiliation wicri:level="4"><inist:fA14 i1="02"><s1>Faculty of Information Science and Electrical Engineering, Kyushu University, Motooka 744</s1>
<s2>Nishi-ku, Fukuoka 819-0395</s2>
<s3>JPN</s3>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>Japon</country>
<wicri:noRegion>Nishi-ku, Fukuoka 819-0395</wicri:noRegion>
<orgName type="university">Université de Kyūshū</orgName>
<placeName><settlement type="city">Fukuoka</settlement>
<region type="province">Kyūshū</region>
<region type="prefecture">Préfecture de Fukuoka</region>
</placeName>
</affiliation>
</author>
<author><name sortKey="Suzuki, Masakazu" sort="Suzuki, Masakazu" uniqKey="Suzuki M" first="Masakazu" last="Suzuki">Masakazu Suzuki (mathématicien)</name>
<affiliation wicri:level="4"><inist:fA14 i1="01"><s1>Faculty of Mathematics, Kyushu University, Hakozaki 6-10-1</s1>
<s2>Higashi-ku, Fukuoka 812-8581</s2>
<s3>JPN</s3>
<sZ>1 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Japon</country>
<wicri:noRegion>Higashi-ku, Fukuoka 812-8581</wicri:noRegion>
<orgName type="university">Université de Kyūshū</orgName>
<placeName><settlement type="city">Fukuoka</settlement>
<region type="province">Kyūshū</region>
<region type="prefecture">Préfecture de Fukuoka</region>
</placeName>
<placeName><settlement type="city">Fukuoka</settlement>
<region type="province">Kyūshū</region>
<region type="prefecture">Préfecture de Fukuoka</region>
</placeName>
<orgName type="university" n="3">Université de Kyūshū</orgName>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">INIST</idno>
<idno type="inist">08-0289256</idno>
<date when="2008">2008</date>
<idno type="stanalyst">PASCAL 08-0289256 INIST</idno>
<idno type="RBID">Pascal:08-0289256</idno>
<idno type="wicri:Area/PascalFrancis/Corpus">000279</idno>
<idno type="wicri:Area/PascalFrancis/Curation">000505</idno>
<idno type="wicri:Area/PascalFrancis/Checkpoint">000233</idno>
<idno type="wicri:doubleKey">0167-8655:2008:Malon C:mathematical:symbol:recognition</idno>
<idno type="wicri:Area/Main/Merge">000D23</idno>
<idno type="wicri:Area/Main/Curation">000D11</idno>
<idno type="wicri:Area/Main/Exploration">000D11</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">Mathematical symbol recognition with support vector machines</title>
<author><name sortKey="Malon, Christopher" sort="Malon, Christopher" uniqKey="Malon C" first="Christopher" last="Malon">Christopher Malon</name>
<affiliation wicri:level="4"><inist:fA14 i1="01"><s1>Faculty of Mathematics, Kyushu University, Hakozaki 6-10-1</s1>
<s2>Higashi-ku, Fukuoka 812-8581</s2>
<s3>JPN</s3>
<sZ>1 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Japon</country>
<wicri:noRegion>Higashi-ku, Fukuoka 812-8581</wicri:noRegion>
<orgName type="university">Université de Kyūshū</orgName>
<placeName><settlement type="city">Fukuoka</settlement>
<region type="province">Kyūshū</region>
<region type="prefecture">Préfecture de Fukuoka</region>
</placeName>
</affiliation>
</author>
<author><name sortKey="Uchida, Seiichi" sort="Uchida, Seiichi" uniqKey="Uchida S" first="Seiichi" last="Uchida">Seiichi Uchida</name>
<affiliation wicri:level="4"><inist:fA14 i1="02"><s1>Faculty of Information Science and Electrical Engineering, Kyushu University, Motooka 744</s1>
<s2>Nishi-ku, Fukuoka 819-0395</s2>
<s3>JPN</s3>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>Japon</country>
<wicri:noRegion>Nishi-ku, Fukuoka 819-0395</wicri:noRegion>
<orgName type="university">Université de Kyūshū</orgName>
<placeName><settlement type="city">Fukuoka</settlement>
<region type="province">Kyūshū</region>
<region type="prefecture">Préfecture de Fukuoka</region>
</placeName>
</affiliation>
</author>
<author><name sortKey="Suzuki, Masakazu" sort="Suzuki, Masakazu" uniqKey="Suzuki M" first="Masakazu" last="Suzuki">Masakazu Suzuki (mathématicien)</name>
<affiliation wicri:level="4"><inist:fA14 i1="01"><s1>Faculty of Mathematics, Kyushu University, Hakozaki 6-10-1</s1>
<s2>Higashi-ku, Fukuoka 812-8581</s2>
<s3>JPN</s3>
<sZ>1 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Japon</country>
<wicri:noRegion>Higashi-ku, Fukuoka 812-8581</wicri:noRegion>
<orgName type="university">Université de Kyūshū</orgName>
<placeName><settlement type="city">Fukuoka</settlement>
<region type="province">Kyūshū</region>
<region type="prefecture">Préfecture de Fukuoka</region>
</placeName>
<placeName><settlement type="city">Fukuoka</settlement>
<region type="province">Kyūshū</region>
<region type="prefecture">Préfecture de Fukuoka</region>
</placeName>
<orgName type="university" n="3">Université de Kyūshū</orgName>
</affiliation>
</author>
</analytic>
<series><title level="j" type="main">Pattern recognition letters</title>
<title level="j" type="abbreviated">Pattern recogn. lett.</title>
<idno type="ISSN">0167-8655</idno>
<imprint><date when="2008">2008</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><title level="j" type="main">Pattern recognition letters</title>
<title level="j" type="abbreviated">Pattern recogn. lett.</title>
<idno type="ISSN">0167-8655</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Character recognition</term>
<term>Kernel method</term>
<term>Learning</term>
<term>Optical character recognition</term>
<term>Pattern recognition</term>
<term>Performance evaluation</term>
<term>Signal classification</term>
<term>Support vector machine</term>
<term>Symbol detection</term>
<term>Two dimensional model</term>
</keywords>
<keywords scheme="Pascal" xml:lang="fr"><term>Détection de symboles</term>
<term>Machine vecteur support</term>
<term>Reconnaissance caractère</term>
<term>Modèle 2 dimensions</term>
<term>Apprentissage</term>
<term>Reconnaissance optique caractère</term>
<term>Evaluation performance</term>
<term>Méthode noyau</term>
<term>Classification signal</term>
<term>Reconnaissance forme</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">Single-character recognition of mathematical symbols poses challenges from its two-dimensional pattern, the variety of similar symbols that must be recognized distinctly, the imbalance and paucity of training data available, and the impossibility of final verification through spell check. We investigate the use of support vector machines to improve the classification of InftyReader, a free system for the OCR of mathematical documents. First, we compare the performance of SVM kernels and feature definitions on pairs of letters that InftyReader usually confuses. Second, we describe a successful approach to multi-class classification with SVM, utilizing the ranking of alternatives within InftyReader's confusion clusters. The inclusion of our technique in InftyReader reduces its misrecognition rate by 41%.</div>
</front>
</TEI>
<affiliations><list><country><li>Japon</li>
</country>
<region><li>Kyūshū</li>
<li>Préfecture de Fukuoka</li>
</region>
<settlement><li>Fukuoka</li>
</settlement>
<orgName><li>Université de Kyūshū</li>
</orgName>
</list>
<tree><country name="Japon"><region name="Kyūshū"><name sortKey="Malon, Christopher" sort="Malon, Christopher" uniqKey="Malon C" first="Christopher" last="Malon">Christopher Malon</name>
</region>
<name sortKey="Suzuki, Masakazu" sort="Suzuki, Masakazu" uniqKey="Suzuki M" first="Masakazu" last="Suzuki">Masakazu Suzuki (mathématicien)</name>
<name sortKey="Uchida, Seiichi" sort="Uchida, Seiichi" uniqKey="Uchida S" first="Seiichi" last="Uchida">Seiichi Uchida</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 000D11 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000D11 | 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:08-0289256 |texte= Mathematical symbol recognition with support vector machines }}
This area was generated with Dilib version V0.6.32. |