A background-thinning-based approach for separating and recognizing connected handwritten digit strings
Identifieur interne : 001F26 ( Main/Exploration ); précédent : 001F25; suivant : 001F27A background-thinning-based approach for separating and recognizing connected handwritten digit strings
Auteurs : Zhongkang Lu [Hong Kong] ; Zheru Chi [Hong Kong] ; Wan-Chi Siu [Hong Kong] ; Pengfei Shi [République populaire de Chine]Source :
- Pattern Recognition [ 0031-3203 ] ; 1997.
Abstract
Most algorithms for segmenting connected handwritten digit strings are based on the analysis of the foreground pixel distributions and the features on the upper/lower contours of the image. In this paper, a new approach is presented to segment connected handwritten two-digit strings based on the thinning of background regions. The algorithm first locates several feature points on the background skeleton of a digit image. Possible segmentation paths are then constructed by matching these feature points. With geometric property measures, all the possible segmentation paths are ranked using fuzzy rules generated from a decision-tree approach. Finally, the top ranked segmentation paths are tested one by one by an optimized nearest neighbor classifier until one of these candidates is accepted based on an acceptance criterion. Experimental results on NIST special database 3 show that our approach can achieve a correct classification rate of 92.5% with only 4.7% of digit strings rejected, which compares favorably with the other techniques tested.
Url:
DOI: 10.1016/S0031-3203(98)00123-X
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: 000B44
- to stream Istex, to step Curation: 000B29
- to stream Istex, to step Checkpoint: 001477
- to stream Main, to step Merge: 002035
- to stream Main, to step Curation: 001F26
Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title>A background-thinning-based approach for separating and recognizing connected handwritten digit strings</title><author><name sortKey="Lu, Zhongkang" sort="Lu, Zhongkang" uniqKey="Lu Z" first="Zhongkang" last="Lu">Zhongkang Lu</name></author><author><name sortKey="Chi, Zheru" sort="Chi, Zheru" uniqKey="Chi Z" first="Zheru" last="Chi">Zheru Chi</name></author><author><name sortKey="Siu, Wan Chi" sort="Siu, Wan Chi" uniqKey="Siu W" first="Wan-Chi" last="Siu">Wan-Chi Siu</name></author><author><name sortKey="Shi, Pengfei" sort="Shi, Pengfei" uniqKey="Shi P" first="Pengfei" last="Shi">Pengfei Shi</name></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:639BFA6F63CEE00F51A5955B7B0D00F37C5430FE</idno><date when="1999" year="1999">1999</date><idno type="doi">10.1016/S0031-3203(98)00123-X</idno><idno type="url">https://api.istex.fr/document/639BFA6F63CEE00F51A5955B7B0D00F37C5430FE/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">000B44</idno><idno type="wicri:Area/Istex/Curation">000B29</idno><idno type="wicri:Area/Istex/Checkpoint">001477</idno><idno type="wicri:doubleKey">0031-3203:1999:Lu Z:a:background:thinning</idno><idno type="wicri:Area/Main/Merge">002035</idno><idno type="wicri:Area/Main/Curation">001F26</idno><idno type="wicri:Area/Main/Exploration">001F26</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a">A background-thinning-based approach for separating and recognizing connected handwritten digit strings</title><author><name sortKey="Lu, Zhongkang" sort="Lu, Zhongkang" uniqKey="Lu Z" first="Zhongkang" last="Lu">Zhongkang Lu</name><affiliation wicri:level="1"><country xml:lang="fr">Hong Kong</country><wicri:regionArea>Department of Electronic Engineering, The Hong Kong Polytechnic University, Hung Hom, Kowloon</wicri:regionArea><wicri:noRegion>Kowloon</wicri:noRegion></affiliation></author><author><name sortKey="Chi, Zheru" sort="Chi, Zheru" uniqKey="Chi Z" first="Zheru" last="Chi">Zheru Chi</name><affiliation wicri:level="1"><country xml:lang="fr">Hong Kong</country><wicri:regionArea>Department of Electronic Engineering, The Hong Kong Polytechnic University, Hung Hom, Kowloon</wicri:regionArea><wicri:noRegion>Kowloon</wicri:noRegion></affiliation></author><author><name sortKey="Siu, Wan Chi" sort="Siu, Wan Chi" uniqKey="Siu W" first="Wan-Chi" last="Siu">Wan-Chi Siu</name><affiliation wicri:level="1"><country xml:lang="fr">Hong Kong</country><wicri:regionArea>Department of Electronic Engineering, The Hong Kong Polytechnic University, Hung Hom, Kowloon</wicri:regionArea><wicri:noRegion>Kowloon</wicri:noRegion></affiliation></author><author><name sortKey="Shi, Pengfei" sort="Shi, Pengfei" uniqKey="Shi P" first="Pengfei" last="Shi">Pengfei Shi</name><affiliation wicri:level="1"><country xml:lang="fr">République populaire de Chine</country><wicri:regionArea>Institute of Pattern Recognition & Image Processing, Shanghai Jiaotong University, Shanghai 200030</wicri:regionArea><wicri:noRegion>Shanghai 200030</wicri:noRegion></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="1997">1997</date><biblScope unit="volume">32</biblScope><biblScope unit="issue">6</biblScope><biblScope unit="page" from="921">921</biblScope><biblScope unit="page" to="933">933</biblScope></imprint><idno type="ISSN">0031-3203</idno></series><idno type="istex">639BFA6F63CEE00F51A5955B7B0D00F37C5430FE</idno><idno type="DOI">10.1016/S0031-3203(98)00123-X</idno><idno type="PII">S0031-3203(98)00123-X</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">Most algorithms for segmenting connected handwritten digit strings are based on the analysis of the foreground pixel distributions and the features on the upper/lower contours of the image. In this paper, a new approach is presented to segment connected handwritten two-digit strings based on the thinning of background regions. The algorithm first locates several feature points on the background skeleton of a digit image. Possible segmentation paths are then constructed by matching these feature points. With geometric property measures, all the possible segmentation paths are ranked using fuzzy rules generated from a decision-tree approach. Finally, the top ranked segmentation paths are tested one by one by an optimized nearest neighbor classifier until one of these candidates is accepted based on an acceptance criterion. Experimental results on NIST special database 3 show that our approach can achieve a correct classification rate of 92.5% with only 4.7% of digit strings rejected, which compares favorably with the other techniques tested.</div></front></TEI><affiliations><list><country><li>Hong Kong</li><li>République populaire de Chine</li></country></list><tree><country name="Hong Kong"><noRegion><name sortKey="Lu, Zhongkang" sort="Lu, Zhongkang" uniqKey="Lu Z" first="Zhongkang" last="Lu">Zhongkang Lu</name></noRegion><name sortKey="Chi, Zheru" sort="Chi, Zheru" uniqKey="Chi Z" first="Zheru" last="Chi">Zheru Chi</name><name sortKey="Siu, Wan Chi" sort="Siu, Wan Chi" uniqKey="Siu W" first="Wan-Chi" last="Siu">Wan-Chi Siu</name></country><country name="République populaire de Chine"><noRegion><name sortKey="Shi, Pengfei" sort="Shi, Pengfei" uniqKey="Shi P" first="Pengfei" last="Shi">Pengfei Shi</name></noRegion></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 001F26 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 001F26 | 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= ISTEX:639BFA6F63CEE00F51A5955B7B0D00F37C5430FE
|texte= A background-thinning-based approach for separating and recognizing connected handwritten digit strings
}}
| This area was generated with Dilib version V0.6.32. |  |