Metric Rectification of Curved Document Images
Identifieur interne : 000308 ( Main/Exploration ); précédent : 000307; suivant : 000309Metric Rectification of Curved Document Images
Auteurs : GAOFENG MENG [République populaire de Chine] ; CHUNHONG PAN [République populaire de Chine] ; SHIMING XIANG [République populaire de Chine] ; JIANGYONG DUAN [République populaire de Chine] ; NANNING ZHENG [République populaire de Chine]Source :
- IEEE transactions on pattern analysis and machine intelligence [ 0162-8828 ] ; 2012.
Descripteurs français
- Pascal (Inist)
- Courbure, Restauration image, Texte, Vision ordinateur, Reconnaissance caractère, Reconnaissance optique caractère, Analyse documentaire, Analyse image, Traitement image, Formation image, Traitement image stéréoscopique, Vision stéréoscopique, Métrique, Rectification, Isométrie, Forme courbe, Projection perspective, Modélisation, Rapport aspect, Correction erreur, Angle observation, Efficacité, Gauchissement, ..
English descriptors
- KwdEn :
- Aspect ratio, Character recognition, Computer vision, Curvature, Curved shape, Document analysis, Efficiency, Error correction, Image analysis, Image processing, Image restoration, Imaging, Isometry, Metric, Modeling, Optical character recognition, Perspective projection, Rectification, Stereo image processing, Stereopsis, Text, Viewing angle, Warping.
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.
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream PascalFrancis, to step Corpus: 000068
- to stream PascalFrancis, to step Curation: 000700
- to stream PascalFrancis, to step Checkpoint: 000070
- to stream Main, to step Merge: 000311
- to stream Main, to step Curation: 000308
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">Metric Rectification of Curved Document Images</title>
<author><name sortKey="Gaofeng Meng" sort="Gaofeng Meng" uniqKey="Gaofeng Meng" last="Gaofeng Meng">GAOFENG MENG</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Zhonggunncun East Road, No. 95</s1>
<s2>Haidian District, Beijing 100190</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
<sZ>4 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
<placeName><settlement type="city">Pékin</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Chunhong Pan" sort="Chunhong Pan" uniqKey="Chunhong Pan" last="Chunhong Pan">CHUNHONG PAN</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Zhonggunncun East Road, No. 95</s1>
<s2>Haidian District, Beijing 100190</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
<sZ>4 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
<placeName><settlement type="city">Pékin</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Shiming Xiang" sort="Shiming Xiang" uniqKey="Shiming Xiang" last="Shiming Xiang">SHIMING XIANG</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Zhonggunncun East Road, No. 95</s1>
<s2>Haidian District, Beijing 100190</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
<sZ>4 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
<placeName><settlement type="city">Pékin</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Jiangyong Duan" sort="Jiangyong Duan" uniqKey="Jiangyong Duan" last="Jiangyong Duan">JIANGYONG DUAN</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Zhonggunncun East Road, No. 95</s1>
<s2>Haidian District, Beijing 100190</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
<sZ>4 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
<placeName><settlement type="city">Pékin</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Nanning Zheng" sort="Nanning Zheng" uniqKey="Nanning Zheng" last="Nanning Zheng">NANNING ZHENG</name>
<affiliation wicri:level="1"><inist:fA14 i1="02"><s1>Institute of Artificial Intelligence and Robotics, Xi'an Jiaotong University</s1>
<s2>Xi'an 710049</s2>
<s3>CHN</s3>
<sZ>5 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
<wicri:noRegion>Xi'an 710049</wicri:noRegion>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">INIST</idno>
<idno type="inist">13-0020849</idno>
<date when="2012">2012</date>
<idno type="stanalyst">PASCAL 13-0020849 INIST</idno>
<idno type="RBID">Pascal:13-0020849</idno>
<idno type="wicri:Area/PascalFrancis/Corpus">000068</idno>
<idno type="wicri:Area/PascalFrancis/Curation">000700</idno>
<idno type="wicri:Area/PascalFrancis/Checkpoint">000070</idno>
<idno type="wicri:doubleKey">0162-8828:2012:Gaofeng Meng:metric:rectification:of</idno>
<idno type="wicri:Area/Main/Merge">000311</idno>
<idno type="wicri:Area/Main/Curation">000308</idno>
<idno type="wicri:Area/Main/Exploration">000308</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">Metric Rectification of Curved Document Images</title>
<author><name sortKey="Gaofeng Meng" sort="Gaofeng Meng" uniqKey="Gaofeng Meng" last="Gaofeng Meng">GAOFENG MENG</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Zhonggunncun East Road, No. 95</s1>
<s2>Haidian District, Beijing 100190</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
<sZ>4 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
<placeName><settlement type="city">Pékin</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Chunhong Pan" sort="Chunhong Pan" uniqKey="Chunhong Pan" last="Chunhong Pan">CHUNHONG PAN</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Zhonggunncun East Road, No. 95</s1>
<s2>Haidian District, Beijing 100190</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
<sZ>4 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
<placeName><settlement type="city">Pékin</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Shiming Xiang" sort="Shiming Xiang" uniqKey="Shiming Xiang" last="Shiming Xiang">SHIMING XIANG</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Zhonggunncun East Road, No. 95</s1>
<s2>Haidian District, Beijing 100190</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
<sZ>4 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
<placeName><settlement type="city">Pékin</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Jiangyong Duan" sort="Jiangyong Duan" uniqKey="Jiangyong Duan" last="Jiangyong Duan">JIANGYONG DUAN</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, Zhonggunncun East Road, No. 95</s1>
<s2>Haidian District, Beijing 100190</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
<sZ>4 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
<placeName><settlement type="city">Pékin</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Nanning Zheng" sort="Nanning Zheng" uniqKey="Nanning Zheng" last="Nanning Zheng">NANNING ZHENG</name>
<affiliation wicri:level="1"><inist:fA14 i1="02"><s1>Institute of Artificial Intelligence and Robotics, Xi'an Jiaotong University</s1>
<s2>Xi'an 710049</s2>
<s3>CHN</s3>
<sZ>5 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
<wicri:noRegion>Xi'an 710049</wicri:noRegion>
</affiliation>
</author>
</analytic>
<series><title level="j" type="main">IEEE transactions on pattern analysis and machine intelligence</title>
<title level="j" type="abbreviated">IEEE trans. pattern anal. mach. intell.</title>
<idno type="ISSN">0162-8828</idno>
<imprint><date when="2012">2012</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><title level="j" type="main">IEEE transactions on pattern analysis and machine intelligence</title>
<title level="j" type="abbreviated">IEEE trans. pattern anal. mach. intell.</title>
<idno type="ISSN">0162-8828</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Aspect ratio</term>
<term>Character recognition</term>
<term>Computer vision</term>
<term>Curvature</term>
<term>Curved shape</term>
<term>Document analysis</term>
<term>Efficiency</term>
<term>Error correction</term>
<term>Image analysis</term>
<term>Image processing</term>
<term>Image restoration</term>
<term>Imaging</term>
<term>Isometry</term>
<term>Metric</term>
<term>Modeling</term>
<term>Optical character recognition</term>
<term>Perspective projection</term>
<term>Rectification</term>
<term>Stereo image processing</term>
<term>Stereopsis</term>
<term>Text</term>
<term>Viewing angle</term>
<term>Warping</term>
</keywords>
<keywords scheme="Pascal" xml:lang="fr"><term>Courbure</term>
<term>Restauration image</term>
<term>Texte</term>
<term>Vision ordinateur</term>
<term>Reconnaissance caractère</term>
<term>Reconnaissance optique caractère</term>
<term>Analyse documentaire</term>
<term>Analyse image</term>
<term>Traitement image</term>
<term>Formation image</term>
<term>Traitement image stéréoscopique</term>
<term>Vision stéréoscopique</term>
<term>Métrique</term>
<term>Rectification</term>
<term>Isométrie</term>
<term>Forme courbe</term>
<term>Projection perspective</term>
<term>Modélisation</term>
<term>Rapport aspect</term>
<term>Correction erreur</term>
<term>Angle observation</term>
<term>Efficacité</term>
<term>Gauchissement</term>
<term>.</term>
</keywords>
</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>
<affiliations><list><country><li>République populaire de Chine</li>
</country>
<settlement><li>Pékin</li>
</settlement>
</list>
<tree><country name="République populaire de Chine"><noRegion><name sortKey="Gaofeng Meng" sort="Gaofeng Meng" uniqKey="Gaofeng Meng" last="Gaofeng Meng">GAOFENG MENG</name>
</noRegion>
<name sortKey="Chunhong Pan" sort="Chunhong Pan" uniqKey="Chunhong Pan" last="Chunhong Pan">CHUNHONG PAN</name>
<name sortKey="Jiangyong Duan" sort="Jiangyong Duan" uniqKey="Jiangyong Duan" last="Jiangyong Duan">JIANGYONG DUAN</name>
<name sortKey="Nanning Zheng" sort="Nanning Zheng" uniqKey="Nanning Zheng" last="Nanning Zheng">NANNING ZHENG</name>
<name sortKey="Shiming Xiang" sort="Shiming Xiang" uniqKey="Shiming Xiang" last="Shiming Xiang">SHIMING XIANG</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 000308 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000308 | 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:13-0020849 |texte= Metric Rectification of Curved Document Images }}
This area was generated with Dilib version V0.6.32. |