Metric rectification of curved document images.
Identifieur interne : 000028 ( PubMed/Checkpoint ); précédent : 000027; suivant : 000029Metric rectification of curved document images.
Auteurs : Gaofeng Meng [République populaire de Chine] ; Chunhong Pan ; Shiming Xiang ; Jiangyong Duan ; Nanning ZhengSource :
- IEEE transactions on pattern analysis and machine intelligence [ 1939-3539 ] ; 2012.
Abstract
In this paper, we propose a metric rectification method to restore an image from a single camera-captured document image. The core idea is to construct an isometric image mesh by exploiting the geometry of page surface and camera. Our method uses a general cylindrical surface (GCS) to model the curved page shape. Under a few proper assumptions, the printed horizontal text lines are shown to be line convergent symmetric. This property is then used to constrain the estimation of various model parameters under perspective projection. We also introduce a paraperspective projection to approximate the nonlinear perspective projection. A set of close-form formulas is thus derived for the estimate of GCS directrix and document aspect ratio. Our method provides a straightforward framework for image metric rectification. It is insensitive to camera positions, viewing angles, and the shapes of document pages. To evaluate the proposed method, we implemented comprehensive experiments on both synthetic and real-captured images. The results demonstrate the efficiency of our method. We also carried out a comparative experiment on the public CBDAR2007 data set. The experimental results show that our method outperforms the state-of-the-art methods in terms of OCR accuracy and rectification errors.
DOI: 10.1109/TPAMI.2011.151
PubMed: 21808093
Affiliations:
Links toward previous steps (curation, corpus...)
Links to Exploration step
pubmed:21808093Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Metric rectification of curved document images.</title><author><name sortKey="Meng, Gaofeng" sort="Meng, Gaofeng" uniqKey="Meng G" first="Gaofeng" last="Meng">Gaofeng Meng</name><affiliation wicri:level="1"><nlm:affiliation>National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Zhongguancun East Road, No. 95, Haidian District, Beijing 100190, P.R. China. gfmeng@nlpr.ia.ac.cn</nlm:affiliation><country wicri:rule="url">République populaire de Chine</country></affiliation></author><author><name sortKey="Pan, Chunhong" sort="Pan, Chunhong" uniqKey="Pan C" first="Chunhong" last="Pan">Chunhong Pan</name></author><author><name sortKey="Xiang, Shiming" sort="Xiang, Shiming" uniqKey="Xiang S" first="Shiming" last="Xiang">Shiming Xiang</name></author><author><name sortKey="Duan, Jiangyong" sort="Duan, Jiangyong" uniqKey="Duan J" first="Jiangyong" last="Duan">Jiangyong Duan</name></author><author><name sortKey="Zheng, Nanning" sort="Zheng, Nanning" uniqKey="Zheng N" first="Nanning" last="Zheng">Nanning Zheng</name></author></titleStmt><publicationStmt><idno type="wicri:source">PubMed</idno><date when="2012">2012</date><idno type="doi">10.1109/TPAMI.2011.151</idno><idno type="RBID">pubmed:21808093</idno><idno type="pmid">21808093</idno><idno type="wicri:Area/PubMed/Corpus">000035</idno><idno type="wicri:Area/PubMed/Curation">000035</idno><idno type="wicri:Area/PubMed/Checkpoint">000035</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">Metric rectification of curved document images.</title><author><name sortKey="Meng, Gaofeng" sort="Meng, Gaofeng" uniqKey="Meng G" first="Gaofeng" last="Meng">Gaofeng Meng</name><affiliation wicri:level="1"><nlm:affiliation>National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Zhongguancun East Road, No. 95, Haidian District, Beijing 100190, P.R. China. gfmeng@nlpr.ia.ac.cn</nlm:affiliation><country wicri:rule="url">République populaire de Chine</country></affiliation></author><author><name sortKey="Pan, Chunhong" sort="Pan, Chunhong" uniqKey="Pan C" first="Chunhong" last="Pan">Chunhong Pan</name></author><author><name sortKey="Xiang, Shiming" sort="Xiang, Shiming" uniqKey="Xiang S" first="Shiming" last="Xiang">Shiming Xiang</name></author><author><name sortKey="Duan, Jiangyong" sort="Duan, Jiangyong" uniqKey="Duan J" first="Jiangyong" last="Duan">Jiangyong Duan</name></author><author><name sortKey="Zheng, Nanning" sort="Zheng, Nanning" uniqKey="Zheng N" first="Nanning" last="Zheng">Nanning Zheng</name></author></analytic><series><title level="j">IEEE transactions on pattern analysis and machine intelligence</title><idno type="eISSN">1939-3539</idno><imprint><date when="2012" type="published">2012</date></imprint></series></biblStruct></sourceDesc></fileDesc><profileDesc><textClass></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">In this paper, we propose a metric rectification method to restore an image from a single camera-captured document image. The core idea is to construct an isometric image mesh by exploiting the geometry of page surface and camera. Our method uses a general cylindrical surface (GCS) to model the curved page shape. Under a few proper assumptions, the printed horizontal text lines are shown to be line convergent symmetric. This property is then used to constrain the estimation of various model parameters under perspective projection. We also introduce a paraperspective projection to approximate the nonlinear perspective projection. A set of close-form formulas is thus derived for the estimate of GCS directrix and document aspect ratio. Our method provides a straightforward framework for image metric rectification. It is insensitive to camera positions, viewing angles, and the shapes of document pages. To evaluate the proposed method, we implemented comprehensive experiments on both synthetic and real-captured images. The results demonstrate the efficiency of our method. We also carried out a comparative experiment on the public CBDAR2007 data set. The experimental results show that our method outperforms the state-of-the-art methods in terms of OCR accuracy and rectification errors.</div></front></TEI><pubmed><MedlineCitation Owner="NLM" Status="PubMed-not-MEDLINE"><PMID Version="1">21808093</PMID><DateCreated><Year>2012</Year><Month>05</Month><Day>31</Day></DateCreated><DateCompleted><Year>2012</Year><Month>09</Month><Day>10</Day></DateCompleted><Article PubModel="Print"><Journal><ISSN IssnType="Electronic">1939-3539</ISSN><JournalIssue CitedMedium="Internet"><Volume>34</Volume><Issue>4</Issue><PubDate><Year>2012</Year><Month>Apr</Month></PubDate></JournalIssue><Title>IEEE transactions on pattern analysis and machine intelligence</Title><ISOAbbreviation>IEEE Trans Pattern Anal Mach Intell</ISOAbbreviation></Journal><ArticleTitle>Metric rectification of curved document images.</ArticleTitle><Pagination><MedlinePgn>707-22</MedlinePgn></Pagination><ELocationID EIdType="doi" ValidYN="Y">10.1109/TPAMI.2011.151</ELocationID><Abstract><AbstractText>In this paper, we propose a metric rectification method to restore an image from a single camera-captured document image. The core idea is to construct an isometric image mesh by exploiting the geometry of page surface and camera. Our method uses a general cylindrical surface (GCS) to model the curved page shape. Under a few proper assumptions, the printed horizontal text lines are shown to be line convergent symmetric. This property is then used to constrain the estimation of various model parameters under perspective projection. We also introduce a paraperspective projection to approximate the nonlinear perspective projection. A set of close-form formulas is thus derived for the estimate of GCS directrix and document aspect ratio. Our method provides a straightforward framework for image metric rectification. It is insensitive to camera positions, viewing angles, and the shapes of document pages. To evaluate the proposed method, we implemented comprehensive experiments on both synthetic and real-captured images. The results demonstrate the efficiency of our method. We also carried out a comparative experiment on the public CBDAR2007 data set. The experimental results show that our method outperforms the state-of-the-art methods in terms of OCR accuracy and rectification errors.</AbstractText></Abstract><AuthorList CompleteYN="Y"><Author ValidYN="Y"><LastName>Meng</LastName><ForeName>Gaofeng</ForeName><Initials>G</Initials><AffiliationInfo><Affiliation>National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Zhongguancun East Road, No. 95, Haidian District, Beijing 100190, P.R. China. gfmeng@nlpr.ia.ac.cn</Affiliation></AffiliationInfo></Author><Author ValidYN="Y"><LastName>Pan</LastName><ForeName>Chunhong</ForeName><Initials>C</Initials></Author><Author ValidYN="Y"><LastName>Xiang</LastName><ForeName>Shiming</ForeName><Initials>S</Initials></Author><Author ValidYN="Y"><LastName>Duan</LastName><ForeName>Jiangyong</ForeName><Initials>J</Initials></Author><Author ValidYN="Y"><LastName>Zheng</LastName><ForeName>Nanning</ForeName><Initials>N</Initials></Author></AuthorList><Language>eng</Language><PublicationTypeList><PublicationType UI="D016428">Journal Article</PublicationType><PublicationType UI="D013485">Research Support, Non-U.S. Gov't</PublicationType></PublicationTypeList></Article><MedlineJournalInfo><Country>United States</Country><MedlineTA>IEEE Trans Pattern Anal Mach Intell</MedlineTA><NlmUniqueID>9885960</NlmUniqueID><ISSNLinking>0098-5589</ISSNLinking></MedlineJournalInfo></MedlineCitation><PubmedData><History><PubMedPubDate PubStatus="entrez"><Year>2011</Year><Month>8</Month><Day>3</Day><Hour>6</Hour><Minute>0</Minute></PubMedPubDate><PubMedPubDate PubStatus="pubmed"><Year>2011</Year><Month>8</Month><Day>3</Day><Hour>6</Hour><Minute>0</Minute></PubMedPubDate><PubMedPubDate PubStatus="medline"><Year>2011</Year><Month>8</Month><Day>3</Day><Hour>6</Hour><Minute>1</Minute></PubMedPubDate></History><PublicationStatus>ppublish</PublicationStatus><ArticleIdList><ArticleId IdType="doi">10.1109/TPAMI.2011.151</ArticleId><ArticleId IdType="pubmed">21808093</ArticleId></ArticleIdList></PubmedData></pubmed><affiliations><list><country><li>République populaire de Chine</li></country></list><tree><noCountry><name sortKey="Duan, Jiangyong" sort="Duan, Jiangyong" uniqKey="Duan J" first="Jiangyong" last="Duan">Jiangyong Duan</name><name sortKey="Pan, Chunhong" sort="Pan, Chunhong" uniqKey="Pan C" first="Chunhong" last="Pan">Chunhong Pan</name><name sortKey="Xiang, Shiming" sort="Xiang, Shiming" uniqKey="Xiang S" first="Shiming" last="Xiang">Shiming Xiang</name><name sortKey="Zheng, Nanning" sort="Zheng, Nanning" uniqKey="Zheng N" first="Nanning" last="Zheng">Nanning Zheng</name></noCountry><country name="République populaire de Chine"><noRegion><name sortKey="Meng, Gaofeng" sort="Meng, Gaofeng" uniqKey="Meng G" first="Gaofeng" last="Meng">Gaofeng Meng</name></noRegion></country></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/PubMed/Checkpoint
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000028 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/PubMed/Checkpoint/biblio.hfd -nk 000028 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Ticri/CIDE
|area= OcrV1
|flux= PubMed
|étape= Checkpoint
|type= RBID
|clé= pubmed:21808093
|texte= Metric rectification of curved document images.
}}
Pour générer des pages wiki
HfdIndexSelect -h $EXPLOR_AREA/Data/PubMed/Checkpoint/RBID.i -Sk "pubmed:21808093" \
| HfdSelect -Kh $EXPLOR_AREA/Data/PubMed/Checkpoint/biblio.hfd \
| NlmPubMed2Wicri -a OcrV1
| This area was generated with Dilib version V0.6.32. |  |