An Extended Learning Vector Quantization Algorithm Aiming at Recognition-Based Character Segmentation
Identifieur interne : 001134 ( Main/Merge ); précédent : 001133; suivant : 001135An Extended Learning Vector Quantization Algorithm Aiming at Recognition-Based Character Segmentation
Auteurs : Lei Xu [République populaire de Chine] ; Bai-Hua Xiao [République populaire de Chine] ; Chun-Heng Wang [République populaire de Chine] ; Ru-Wei Dai [République populaire de Chine]Source :
- Lecture Notes in Control and Information Sciences [ 0170-8643 ] ; 2006.
Abstract
Abstract: Recognition-based segmentation strategies have greatly improved the performance of optical character recognition systems. The key issue of these strategies is to design a classifier that can provide accurate rejection information. Many learning algorithms, such as GLVQ and H2M-LVQ, are not suitable for large category sets and multiple prototypes. More seriously, they often suffer from local minimum state and overtraining. In this paper, we propose an extended learning vector quantization algorithm which can efficiently train the nearest prototype classifier with negative samples. The cost function is based on multiple confusable prototype-pairs so that our algorithm is insensitive to initialization. We also introduce the criterion of safe zone to avoid overtraining. Experimental results show that the classifier trained by our proposed method can achieve good recognition performance and can provide accurate rejection information for segmentation.
Url:
DOI: 10.1007/978-3-540-37258-5_14
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: 000B62
- to stream Istex, to step Curation: 000B47
- to stream Istex, to step Checkpoint: 000A90
Links to Exploration step
ISTEX:F7F0249220DF6619A1A528B89703C417C6AEFD8DLe document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">An Extended Learning Vector Quantization Algorithm Aiming at Recognition-Based Character Segmentation</title><author><name sortKey="Xu, Lei" sort="Xu, Lei" uniqKey="Xu L" first="Lei" last="Xu">Lei Xu</name></author><author><name sortKey="Xiao, Bai Hua" sort="Xiao, Bai Hua" uniqKey="Xiao B" first="Bai-Hua" last="Xiao">Bai-Hua Xiao</name></author><author><name sortKey="Wang, Chun Heng" sort="Wang, Chun Heng" uniqKey="Wang C" first="Chun-Heng" last="Wang">Chun-Heng Wang</name></author><author><name sortKey="Dai, Ru Wei" sort="Dai, Ru Wei" uniqKey="Dai R" first="Ru-Wei" last="Dai">Ru-Wei Dai</name></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:F7F0249220DF6619A1A528B89703C417C6AEFD8D</idno><date when="2006" year="2006">2006</date><idno type="doi">10.1007/978-3-540-37258-5_14</idno><idno type="url">https://api.istex.fr/document/F7F0249220DF6619A1A528B89703C417C6AEFD8D/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">000B62</idno><idno type="wicri:Area/Istex/Curation">000B47</idno><idno type="wicri:Area/Istex/Checkpoint">000A90</idno><idno type="wicri:doubleKey">0170-8643:2006:Xu L:an:extended:learning</idno><idno type="wicri:Area/Main/Merge">001134</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">An Extended Learning Vector Quantization Algorithm Aiming at Recognition-Based Character Segmentation</title><author><name sortKey="Xu, Lei" sort="Xu, Lei" uniqKey="Xu L" first="Lei" last="Xu">Lei Xu</name><affiliation wicri:level="3"><country xml:lang="fr" wicri:curation="lc">République populaire de Chine</country><wicri:regionArea>Laboratory of Complex System and Intelligent Science Institute of Automation, Chinese Academy of Sciences, Zhongguancun East Rd, No.95, 100080, Beijing</wicri:regionArea><placeName><settlement type="city">Pékin</settlement></placeName></affiliation><affiliation wicri:level="1"><country wicri:rule="url">République populaire de Chine</country></affiliation></author><author><name sortKey="Xiao, Bai Hua" sort="Xiao, Bai Hua" uniqKey="Xiao B" first="Bai-Hua" last="Xiao">Bai-Hua Xiao</name><affiliation wicri:level="3"><country xml:lang="fr" wicri:curation="lc">République populaire de Chine</country><wicri:regionArea>Laboratory of Complex System and Intelligent Science Institute of Automation, Chinese Academy of Sciences, Zhongguancun East Rd, No.95, 100080, Beijing</wicri:regionArea><placeName><settlement type="city">Pékin</settlement></placeName></affiliation></author><author><name sortKey="Wang, Chun Heng" sort="Wang, Chun Heng" uniqKey="Wang C" first="Chun-Heng" last="Wang">Chun-Heng Wang</name><affiliation wicri:level="3"><country xml:lang="fr" wicri:curation="lc">République populaire de Chine</country><wicri:regionArea>Laboratory of Complex System and Intelligent Science Institute of Automation, Chinese Academy of Sciences, Zhongguancun East Rd, No.95, 100080, Beijing</wicri:regionArea><placeName><settlement type="city">Pékin</settlement></placeName></affiliation></author><author><name sortKey="Dai, Ru Wei" sort="Dai, Ru Wei" uniqKey="Dai R" first="Ru-Wei" last="Dai">Ru-Wei Dai</name><affiliation wicri:level="3"><country xml:lang="fr" wicri:curation="lc">République populaire de Chine</country><wicri:regionArea>Laboratory of Complex System and Intelligent Science Institute of Automation, Chinese Academy of Sciences, Zhongguancun East Rd, No.95, 100080, Beijing</wicri:regionArea><placeName><settlement type="city">Pékin</settlement></placeName></affiliation></author></analytic><monogr></monogr><series><title level="s">Lecture Notes in Control and Information Sciences</title><imprint><date>2006</date></imprint><idno type="ISSN">0170-8643</idno><idno type="ISSN">0170-8643</idno></series><idno type="istex">F7F0249220DF6619A1A528B89703C417C6AEFD8D</idno><idno type="DOI">10.1007/978-3-540-37258-5_14</idno><idno type="ChapterID">14</idno><idno type="ChapterID">Chap14</idno></biblStruct></sourceDesc><seriesStmt><idno type="ISSN">0170-8643</idno></seriesStmt></fileDesc><profileDesc><textClass></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Abstract: Recognition-based segmentation strategies have greatly improved the performance of optical character recognition systems. The key issue of these strategies is to design a classifier that can provide accurate rejection information. Many learning algorithms, such as GLVQ and H2M-LVQ, are not suitable for large category sets and multiple prototypes. More seriously, they often suffer from local minimum state and overtraining. In this paper, we propose an extended learning vector quantization algorithm which can efficiently train the nearest prototype classifier with negative samples. The cost function is based on multiple confusable prototype-pairs so that our algorithm is insensitive to initialization. We also introduce the criterion of safe zone to avoid overtraining. Experimental results show that the classifier trained by our proposed method can achieve good recognition performance and can provide accurate rejection information for segmentation.</div></front></TEI></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Main/Merge
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 001134 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Merge/biblio.hfd -nk 001134 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Ticri/CIDE
|area= OcrV1
|flux= Main
|étape= Merge
|type= RBID
|clé= ISTEX:F7F0249220DF6619A1A528B89703C417C6AEFD8D
|texte= An Extended Learning Vector Quantization Algorithm Aiming at Recognition-Based Character Segmentation
}}
| This area was generated with Dilib version V0.6.32. |  |