Composition of a dewarped and enhanced document image from two view images.
Identifieur interne : 000815 ( Main/Exploration ); précédent : 000814; suivant : 000816Composition of a dewarped and enhanced document image from two view images.
Auteurs : Hyung Il Koo [Corée du Sud] ; Jinho Kim ; Nam Ik ChoSource :
- IEEE transactions on image processing : a publication of the IEEE Signal Processing Society [ 1057-7149 ] ; 2009.
Abstract
In this paper, we propose an algorithm to compose a geometrically dewarped and visually enhanced image from two document images taken by a digital camera at different angles. Unlike the conventional works that require special equipment or assumptions on the contents of books or complicated image acquisition steps, we estimate the unfolded book or document surface from the corresponding points between two images. For this purpose, the surface and camera matrices are estimated using structure reconstruction, 3-D projection analysis, and random sample consensus-based curve fitting with the cylindrical surface model. Because we do not need any assumption on the contents of books, the proposed method can be applied not only to optical character recognition (OCR), but also to the high-quality digitization of pictures in documents. In addition to the dewarping for a structurally better image, image mosaic is also performed for further improving the visual quality. By finding better parts of images (with less out of focus blur and/or without specular reflections) from either of views, we compose a better image by stitching and blending them. These processes are formulated as energy minimization problems that can be solved using a graph cut method. Experiments on many kinds of book or document images show that the proposed algorithm robustly works and yields visually pleasing results. Also, the OCR rate of the resulting image is comparable to that of document images from a flatbed scanner.
DOI: 10.1109/TIP.2009.2019301
PubMed: 19447710
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream PubMed, to step Corpus: 000043
- to stream PubMed, to step Curation: 000043
- to stream PubMed, to step Checkpoint: 000043
- to stream Ncbi, to step Merge: 000072
- to stream Ncbi, to step Curation: 000072
- to stream Ncbi, to step Checkpoint: 000072
- to stream Main, to step Merge: 000823
- to stream Main, to step Curation: 000815
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Composition of a dewarped and enhanced document image from two view images.</title><author><name sortKey="Koo, Hyung Il" sort="Koo, Hyung Il" uniqKey="Koo H" first="Hyung Il" last="Koo">Hyung Il Koo</name><affiliation wicri:level="4"><nlm:affiliation>Department of Electrical Engineering and Computer Science, Seoul National University, Seoul, Korea. hikoo@ispl.snu.ac.kr</nlm:affiliation><country xml:lang="fr">Corée du Sud</country><wicri:regionArea>Department of Electrical Engineering and Computer Science, Seoul National University, Seoul</wicri:regionArea><placeName><settlement type="city">Séoul</settlement></placeName><orgName type="university">Université nationale de Séoul</orgName></affiliation></author><author><name sortKey="Kim, Jinho" sort="Kim, Jinho" uniqKey="Kim J" first="Jinho" last="Kim">Jinho Kim</name></author><author><name sortKey="Cho, Nam Ik" sort="Cho, Nam Ik" uniqKey="Cho N" first="Nam Ik" last="Cho">Nam Ik Cho</name></author></titleStmt><publicationStmt><idno type="wicri:source">PubMed</idno><date when="2009">2009</date><idno type="doi">10.1109/TIP.2009.2019301</idno><idno type="RBID">pubmed:19447710</idno><idno type="pmid">19447710</idno><idno type="wicri:Area/PubMed/Corpus">000043</idno><idno type="wicri:Area/PubMed/Curation">000043</idno><idno type="wicri:Area/PubMed/Checkpoint">000043</idno><idno type="wicri:Area/Ncbi/Merge">000072</idno><idno type="wicri:Area/Ncbi/Curation">000072</idno><idno type="wicri:Area/Ncbi/Checkpoint">000072</idno><idno type="wicri:doubleKey">1057-7149:2009:Koo H:composition:of:a</idno><idno type="wicri:Area/Main/Merge">000823</idno><idno type="wicri:Area/Main/Curation">000815</idno><idno type="wicri:Area/Main/Exploration">000815</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">Composition of a dewarped and enhanced document image from two view images.</title><author><name sortKey="Koo, Hyung Il" sort="Koo, Hyung Il" uniqKey="Koo H" first="Hyung Il" last="Koo">Hyung Il Koo</name><affiliation wicri:level="4"><nlm:affiliation>Department of Electrical Engineering and Computer Science, Seoul National University, Seoul, Korea. hikoo@ispl.snu.ac.kr</nlm:affiliation><country xml:lang="fr">Corée du Sud</country><wicri:regionArea>Department of Electrical Engineering and Computer Science, Seoul National University, Seoul</wicri:regionArea><placeName><settlement type="city">Séoul</settlement></placeName><orgName type="university">Université nationale de Séoul</orgName></affiliation></author><author><name sortKey="Kim, Jinho" sort="Kim, Jinho" uniqKey="Kim J" first="Jinho" last="Kim">Jinho Kim</name></author><author><name sortKey="Cho, Nam Ik" sort="Cho, Nam Ik" uniqKey="Cho N" first="Nam Ik" last="Cho">Nam Ik Cho</name></author></analytic><series><title level="j">IEEE transactions on image processing : a publication of the IEEE Signal Processing Society</title><idno type="ISSN">1057-7149</idno><imprint><date when="2009" type="published">2009</date></imprint></series></biblStruct></sourceDesc></fileDesc><profileDesc><textClass></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">In this paper, we propose an algorithm to compose a geometrically dewarped and visually enhanced image from two document images taken by a digital camera at different angles. Unlike the conventional works that require special equipment or assumptions on the contents of books or complicated image acquisition steps, we estimate the unfolded book or document surface from the corresponding points between two images. For this purpose, the surface and camera matrices are estimated using structure reconstruction, 3-D projection analysis, and random sample consensus-based curve fitting with the cylindrical surface model. Because we do not need any assumption on the contents of books, the proposed method can be applied not only to optical character recognition (OCR), but also to the high-quality digitization of pictures in documents. In addition to the dewarping for a structurally better image, image mosaic is also performed for further improving the visual quality. By finding better parts of images (with less out of focus blur and/or without specular reflections) from either of views, we compose a better image by stitching and blending them. These processes are formulated as energy minimization problems that can be solved using a graph cut method. Experiments on many kinds of book or document images show that the proposed algorithm robustly works and yields visually pleasing results. Also, the OCR rate of the resulting image is comparable to that of document images from a flatbed scanner.</div></front></TEI><affiliations><list><country><li>Corée du Sud</li></country><settlement><li>Séoul</li></settlement><orgName><li>Université nationale de Séoul</li></orgName></list><tree><noCountry><name sortKey="Cho, Nam Ik" sort="Cho, Nam Ik" uniqKey="Cho N" first="Nam Ik" last="Cho">Nam Ik Cho</name><name sortKey="Kim, Jinho" sort="Kim, Jinho" uniqKey="Kim J" first="Jinho" last="Kim">Jinho Kim</name></noCountry><country name="Corée du Sud"><noRegion><name sortKey="Koo, Hyung Il" sort="Koo, Hyung Il" uniqKey="Koo H" first="Hyung Il" last="Koo">Hyung Il Koo</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 000815 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000815 | 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= pubmed:19447710
|texte= Composition of a dewarped and enhanced document image from two view images.
}}
Pour générer des pages wiki
HfdIndexSelect -h $EXPLOR_AREA/Data/Main/Exploration/RBID.i -Sk "pubmed:19447710" \
| HfdSelect -Kh $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd \
| NlmPubMed2Wicri -a OcrV1
| This area was generated with Dilib version V0.6.32. |  |