An MRF Model for Binarization of Natural Scene Text
Identifieur interne : 000335 ( Main/Exploration ); précédent : 000334; suivant : 000336An MRF Model for Binarization of Natural Scene Text
Auteurs : Anand Mishra [Inde] ; Karteek Alahari [France] ; C. V. Jawahar [Inde]Source :
English descriptors
- mix :
- Binarization, GMM, Graph Cut, MRF.
Abstract
Inspired by the success of MRF models for solving object segmentation problems, we formulate the binarization problem in this framework. We represent the pixels in a document image as random variables in an MRF, and introduce a new energy (or cost) function on these variables. Each variable takes a foreground or background label, and the quality of the binarization (or labelling) is determined by the value of the energy function. We minimize the energy function, i.e. find the optimal binarization, using an iterative graph cut scheme. Our model is robust to variations in foreground and background colours as we use a Gaussian Mixture Model in the energy function. In addition, our algorithm is efficient to compute, and adapts to a variety of document images. We show results on word images from the challenging ICDAR 2003 dataset, and compare our performance with previously reported methods. Our approach shows significant improvement in pixel level accuracy as well as OCR accuracy.
Url:
DOI: 10.1109/ICDAR.2011.12
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream Hal, to step Corpus: 000019
- to stream Hal, to step Curation: 000019
- to stream Hal, to step Checkpoint: 000082
- to stream Main, to step Merge: 000340
- to stream Main, to step Curation: 000335
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">An MRF Model for Binarization of Natural Scene Text</title><author><name sortKey="Mishra, Anand" sort="Mishra, Anand" uniqKey="Mishra A" first="Anand" last="Mishra">Anand Mishra</name><affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-21854" status="VALID"><orgName>Center for Visual Information Technology [Hyderabad]</orgName><orgName type="acronym">CVIT</orgName><desc><address><addrLine>Gachibowli, Hyderabad - 500 032,Andhra Pradesh</addrLine><country key="IN"></country></address><ref type="url">http://cvit.iiit.ac.in/</ref></desc><listRelation><relation active="#struct-300171" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-300171" type="direct"><org type="institution" xml:id="struct-300171" status="VALID"><orgName>International Institute of Information Technology, Hyderabad [Hyderabad]</orgName><orgName type="acronym">IIIT-H</orgName><desc><address><addrLine>Gachibowli, Hyderabad 500 032Telangana</addrLine><country key="IN"></country></address><ref type="url">https://www.iiit.ac.in/institute/about</ref></desc></org></tutelle></tutelles></hal:affiliation><country>Inde</country></affiliation></author><author><name sortKey="Alahari, Karteek" sort="Alahari, Karteek" uniqKey="Alahari K" first="Karteek" last="Alahari">Karteek Alahari</name><affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-1315" status="VALID"><orgName>Laboratoire d'informatique de l'école normale supérieure</orgName><orgName type="acronym">LIENS</orgName><desc><address><addrLine>45 Rue d'Ulm 75230 PARIS CEDEX 05</addrLine><country key="FR"></country></address><ref type="url">http://www.di.ens.fr</ref></desc><listRelation><relation name="UMR8548" active="#struct-441569" type="direct"></relation><relation active="#struct-59704" type="direct"></relation></listRelation><tutelles><tutelle name="UMR8548" active="#struct-441569" type="direct"><org type="institution" xml:id="struct-441569" status="VALID"><idno type="IdRef">02636817X</idno><idno type="ISNI">0000000122597504</idno><orgName>Centre National de la Recherche Scientifique</orgName><orgName type="acronym">CNRS</orgName><date type="start">1939-10-19</date><desc><address><country key="FR"></country></address><ref type="url">http://www.cnrs.fr/</ref></desc></org></tutelle><tutelle active="#struct-59704" type="direct"><org type="institution" xml:id="struct-59704" status="VALID"><orgName>École normale supérieure - Paris</orgName><orgName type="acronym">ENS Paris</orgName><desc><address><addrLine>45, Rue d'Ulm - 75230 Paris cedex 05</addrLine><country key="FR"></country></address><ref type="url">http://www.ens.fr</ref></desc></org></tutelle></tutelles></hal:affiliation><country>France</country></affiliation></author><author><name sortKey="Jawahar, C V" sort="Jawahar, C V" uniqKey="Jawahar C" first="C. V." last="Jawahar">C. V. Jawahar</name><affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-21854" status="VALID"><orgName>Center for Visual Information Technology [Hyderabad]</orgName><orgName type="acronym">CVIT</orgName><desc><address><addrLine>Gachibowli, Hyderabad - 500 032,Andhra Pradesh</addrLine><country key="IN"></country></address><ref type="url">http://cvit.iiit.ac.in/</ref></desc><listRelation><relation active="#struct-300171" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-300171" type="direct"><org type="institution" xml:id="struct-300171" status="VALID"><orgName>International Institute of Information Technology, Hyderabad [Hyderabad]</orgName><orgName type="acronym">IIIT-H</orgName><desc><address><addrLine>Gachibowli, Hyderabad 500 032Telangana</addrLine><country key="IN"></country></address><ref type="url">https://www.iiit.ac.in/institute/about</ref></desc></org></tutelle></tutelles></hal:affiliation><country>Inde</country></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">HAL</idno><idno type="RBID">Hal:hal-00817972</idno><idno type="halId">hal-00817972</idno><idno type="halUri">https://hal.inria.fr/hal-00817972</idno><idno type="url">https://hal.inria.fr/hal-00817972</idno><idno type="doi">10.1109/ICDAR.2011.12</idno><date when="2011-09-18">2011-09-18</date><idno type="wicri:Area/Hal/Corpus">000019</idno><idno type="wicri:Area/Hal/Curation">000019</idno><idno type="wicri:Area/Hal/Checkpoint">000082</idno><idno type="wicri:Area/Main/Merge">000340</idno><idno type="wicri:Area/Main/Curation">000335</idno><idno type="wicri:Area/Main/Exploration">000335</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">An MRF Model for Binarization of Natural Scene Text</title><author><name sortKey="Mishra, Anand" sort="Mishra, Anand" uniqKey="Mishra A" first="Anand" last="Mishra">Anand Mishra</name><affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-21854" status="VALID"><orgName>Center for Visual Information Technology [Hyderabad]</orgName><orgName type="acronym">CVIT</orgName><desc><address><addrLine>Gachibowli, Hyderabad - 500 032,Andhra Pradesh</addrLine><country key="IN"></country></address><ref type="url">http://cvit.iiit.ac.in/</ref></desc><listRelation><relation active="#struct-300171" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-300171" type="direct"><org type="institution" xml:id="struct-300171" status="VALID"><orgName>International Institute of Information Technology, Hyderabad [Hyderabad]</orgName><orgName type="acronym">IIIT-H</orgName><desc><address><addrLine>Gachibowli, Hyderabad 500 032Telangana</addrLine><country key="IN"></country></address><ref type="url">https://www.iiit.ac.in/institute/about</ref></desc></org></tutelle></tutelles></hal:affiliation><country>Inde</country></affiliation></author><author><name sortKey="Alahari, Karteek" sort="Alahari, Karteek" uniqKey="Alahari K" first="Karteek" last="Alahari">Karteek Alahari</name><affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-1315" status="VALID"><orgName>Laboratoire d'informatique de l'école normale supérieure</orgName><orgName type="acronym">LIENS</orgName><desc><address><addrLine>45 Rue d'Ulm 75230 PARIS CEDEX 05</addrLine><country key="FR"></country></address><ref type="url">http://www.di.ens.fr</ref></desc><listRelation><relation name="UMR8548" active="#struct-441569" type="direct"></relation><relation active="#struct-59704" type="direct"></relation></listRelation><tutelles><tutelle name="UMR8548" active="#struct-441569" type="direct"><org type="institution" xml:id="struct-441569" status="VALID"><idno type="IdRef">02636817X</idno><idno type="ISNI">0000000122597504</idno><orgName>Centre National de la Recherche Scientifique</orgName><orgName type="acronym">CNRS</orgName><date type="start">1939-10-19</date><desc><address><country key="FR"></country></address><ref type="url">http://www.cnrs.fr/</ref></desc></org></tutelle><tutelle active="#struct-59704" type="direct"><org type="institution" xml:id="struct-59704" status="VALID"><orgName>École normale supérieure - Paris</orgName><orgName type="acronym">ENS Paris</orgName><desc><address><addrLine>45, Rue d'Ulm - 75230 Paris cedex 05</addrLine><country key="FR"></country></address><ref type="url">http://www.ens.fr</ref></desc></org></tutelle></tutelles></hal:affiliation><country>France</country></affiliation></author><author><name sortKey="Jawahar, C V" sort="Jawahar, C V" uniqKey="Jawahar C" first="C. V." last="Jawahar">C. V. Jawahar</name><affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-21854" status="VALID"><orgName>Center for Visual Information Technology [Hyderabad]</orgName><orgName type="acronym">CVIT</orgName><desc><address><addrLine>Gachibowli, Hyderabad - 500 032,Andhra Pradesh</addrLine><country key="IN"></country></address><ref type="url">http://cvit.iiit.ac.in/</ref></desc><listRelation><relation active="#struct-300171" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-300171" type="direct"><org type="institution" xml:id="struct-300171" status="VALID"><orgName>International Institute of Information Technology, Hyderabad [Hyderabad]</orgName><orgName type="acronym">IIIT-H</orgName><desc><address><addrLine>Gachibowli, Hyderabad 500 032Telangana</addrLine><country key="IN"></country></address><ref type="url">https://www.iiit.ac.in/institute/about</ref></desc></org></tutelle></tutelles></hal:affiliation><country>Inde</country></affiliation></author></analytic><idno type="DOI">10.1109/ICDAR.2011.12</idno></biblStruct></sourceDesc></fileDesc><profileDesc><textClass><keywords scheme="mix" xml:lang="en"><term>Binarization</term><term>GMM</term><term>Graph Cut</term><term>MRF</term></keywords></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Inspired by the success of MRF models for solving object segmentation problems, we formulate the binarization problem in this framework. We represent the pixels in a document image as random variables in an MRF, and introduce a new energy (or cost) function on these variables. Each variable takes a foreground or background label, and the quality of the binarization (or labelling) is determined by the value of the energy function. We minimize the energy function, i.e. find the optimal binarization, using an iterative graph cut scheme. Our model is robust to variations in foreground and background colours as we use a Gaussian Mixture Model in the energy function. In addition, our algorithm is efficient to compute, and adapts to a variety of document images. We show results on word images from the challenging ICDAR 2003 dataset, and compare our performance with previously reported methods. Our approach shows significant improvement in pixel level accuracy as well as OCR accuracy.</div></front></TEI><affiliations><list><country><li>France</li><li>Inde</li></country></list><tree><country name="Inde"><noRegion><name sortKey="Mishra, Anand" sort="Mishra, Anand" uniqKey="Mishra A" first="Anand" last="Mishra">Anand Mishra</name></noRegion><name sortKey="Jawahar, C V" sort="Jawahar, C V" uniqKey="Jawahar C" first="C. V." last="Jawahar">C. V. Jawahar</name></country><country name="France"><noRegion><name sortKey="Alahari, Karteek" sort="Alahari, Karteek" uniqKey="Alahari K" first="Karteek" last="Alahari">Karteek Alahari</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 000335 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000335 | 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= Hal:hal-00817972
|texte= An MRF Model for Binarization of Natural Scene Text
}}
| This area was generated with Dilib version V0.6.32. |  |