OCR performance prediction using cross-OCR alignment
Identifieur interne : 000089 ( Hal/Corpus ); précédent : 000088; suivant : 000090OCR performance prediction using cross-OCR alignment
Auteurs : Ahmed Ben Salah ; Jean-Philippe Moreux ; Nicolas Ragot ; Thierry PaquetSource :
English descriptors
- mix :
Abstract
Since 2006 the national library of France (BnF) has developed many mass digitization projects on its collections. The indexation of digital documents on Gallica (the digital library of the BnF) is done through their textual content obtained thanks to service providers that use Optical Character Recognition software (OCR). The modern technologies of OCR achieve good performances on modern documents produced with uniform layout and known fonts. However, for old documents, OCR results are of lower quality. The OCR quality assessment is a real challenge for the BnF. On the one hand, due to the sequential architecture of OCR treatments, the identification of OCR errors sources is intractable. On the other hand, besides the word confidence, no additional quality information is reported in OCR outputs. In this paper, we present a study on OCR performance estimation aiming to control the quality of word transcriptions achieved by OCR. This quality assessment process has to operate without any comparison with ground truthed data. In this respect, our methodology relies on cross alignment of the OCR results with those of a secondary OCR called reference OCR. This secondary OCR provides uncertain but useful information that will be used as uncertain groundtruth. OCR performance is estimated using support vector regression. This predictor uses some global features computed on the cross-alignment results. The experimentations reported show that our estimate describes more faithfully the quality of OCR outputs than average word confidence scores that are computed by OCR. The proposed methodology can be adapted easily to various corpora by tuning the system using a training dataset of documents that have similar properties to those to be treated.
Url:
Links to Exploration step
Hal:hal-01191701Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">OCR performance prediction using cross-OCR alignment</title><author><name sortKey="Ben Salah, Ahmed" sort="Ben Salah, Ahmed" uniqKey="Ben Salah A" first="Ahmed" last="Ben Salah">Ahmed Ben Salah</name><affiliation><hal:affiliation type="laboratory" xml:id="struct-205116" status="VALID"><orgName>Bibliothèque nationale de France, Département de la Conservation</orgName><orgName type="acronym">BnF_DSC</orgName><desc><address><addrLine>Quai François Mauriac, 75706 Paris cedex 13</addrLine><country key="FR"></country></address><ref type="url">http://www.bnf.fr/fr/professionnels/conservation.html</ref></desc><listRelation><relation active="#struct-300057" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-300057" type="direct"><org type="institution" xml:id="struct-300057" status="VALID"><orgName>Bibliothèque Nationale de France</orgName><orgName type="acronym">BNF</orgName><desc><address><country key="FR"></country></address></desc></org></tutelle></tutelles></hal:affiliation></affiliation></author><author><name sortKey="Moreux, Jean Philippe" sort="Moreux, Jean Philippe" uniqKey="Moreux J" first="Jean-Philippe" last="Moreux">Jean-Philippe Moreux</name><affiliation><hal:affiliation type="laboratory" xml:id="struct-205116" status="VALID"><orgName>Bibliothèque nationale de France, Département de la Conservation</orgName><orgName type="acronym">BnF_DSC</orgName><desc><address><addrLine>Quai François Mauriac, 75706 Paris cedex 13</addrLine><country key="FR"></country></address><ref type="url">http://www.bnf.fr/fr/professionnels/conservation.html</ref></desc><listRelation><relation active="#struct-300057" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-300057" type="direct"><org type="institution" xml:id="struct-300057" status="VALID"><orgName>Bibliothèque Nationale de France</orgName><orgName type="acronym">BNF</orgName><desc><address><country key="FR"></country></address></desc></org></tutelle></tutelles></hal:affiliation></affiliation></author><author><name sortKey="Ragot, Nicolas" sort="Ragot, Nicolas" uniqKey="Ragot N" first="Nicolas" last="Ragot">Nicolas Ragot</name><affiliation><hal:affiliation type="laboratory" xml:id="struct-204893" status="VALID"><orgName>Laboratoire d'Informatique de l'Université de Tours</orgName><orgName type="acronym">LI</orgName><desc><address><addrLine>64, Avenue Jean Portalis, 37200 Tours</addrLine><country key="FR"></country></address><ref type="url">http://www.li.univ-tours.fr/</ref></desc><listRelation><relation active="#struct-300408" type="direct"></relation><relation name="EA6300" active="#struct-300298" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-300408" type="direct"><org type="institution" xml:id="struct-300408" status="VALID"><orgName>Polytech'Tours</orgName><desc><address><country key="FR"></country></address></desc></org></tutelle><tutelle name="EA6300" active="#struct-300298" type="direct"><org type="institution" xml:id="struct-300298" status="VALID"><orgName>Université François Rabelais - Tours</orgName><desc><address><addrLine>60 rue du Plat d'Étain, 37020 Tours cedex 1 </addrLine><country key="FR"></country></address><ref type="url">http://www.univ-tours.fr</ref></desc></org></tutelle></tutelles></hal:affiliation></affiliation></author><author><name sortKey="Paquet, Thierry" sort="Paquet, Thierry" uniqKey="Paquet T" first="Thierry" last="Paquet">Thierry Paquet</name><affiliation><hal:affiliation type="laboratory" xml:id="struct-23832" status="VALID"><orgName>Laboratoire d'Informatique, de Traitement de l'Information et des Systèmes</orgName><orgName type="acronym">LITIS</orgName><desc><address><addrLine>Avenue de l'Université UFR des Sciences et Techniques 76800 Saint-Etienne du Rouvray</addrLine><country key="FR"></country></address><ref type="url">http://www.litislab.eu</ref></desc><listRelation><relation active="#struct-300317" type="direct"></relation><relation name="EA4108" active="#struct-300318" type="direct"></relation><relation active="#struct-301288" type="direct"></relation><relation active="#struct-301232" type="indirect"></relation></listRelation><tutelles><tutelle active="#struct-300317" type="direct"><org type="institution" xml:id="struct-300317" status="VALID"><orgName>Université du Havre</orgName><desc><address><country key="FR"></country></address></desc></org></tutelle><tutelle name="EA4108" active="#struct-300318" type="direct"><org type="institution" xml:id="struct-300318" status="VALID"><orgName>Université de Rouen</orgName><desc><address><addrLine> 1 rue Thomas Becket - 76821 Mont-Saint-Aignan</addrLine><country key="FR"></country></address><ref type="url">http://www.univ-rouen.fr/</ref></desc></org></tutelle><tutelle active="#struct-301288" type="direct"><org type="department" xml:id="struct-301288" status="VALID"><orgName>Institut National des Sciences Appliquées - Rouen</orgName><orgName type="acronym">INSA Rouen</orgName><desc><address><country key="FR"></country></address></desc><listRelation><relation active="#struct-301232" type="direct"></relation></listRelation></org></tutelle><tutelle active="#struct-301232" type="indirect"><org type="institution" xml:id="struct-301232" status="VALID"><orgName>Institut National des Sciences Appliquées</orgName><orgName type="acronym">INSA</orgName><desc><address><country key="FR"></country></address></desc></org></tutelle></tutelles></hal:affiliation></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">HAL</idno><idno type="RBID">Hal:hal-01191701</idno><idno type="halId">hal-01191701</idno><idno type="halUri">https://hal.archives-ouvertes.fr/hal-01191701</idno><idno type="url">https://hal.archives-ouvertes.fr/hal-01191701</idno><date when="2015-08-23">2015-08-23</date><idno type="wicri:Area/Hal/Corpus">000089</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">OCR performance prediction using cross-OCR alignment</title><author><name sortKey="Ben Salah, Ahmed" sort="Ben Salah, Ahmed" uniqKey="Ben Salah A" first="Ahmed" last="Ben Salah">Ahmed Ben Salah</name><affiliation><hal:affiliation type="laboratory" xml:id="struct-205116" status="VALID"><orgName>Bibliothèque nationale de France, Département de la Conservation</orgName><orgName type="acronym">BnF_DSC</orgName><desc><address><addrLine>Quai François Mauriac, 75706 Paris cedex 13</addrLine><country key="FR"></country></address><ref type="url">http://www.bnf.fr/fr/professionnels/conservation.html</ref></desc><listRelation><relation active="#struct-300057" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-300057" type="direct"><org type="institution" xml:id="struct-300057" status="VALID"><orgName>Bibliothèque Nationale de France</orgName><orgName type="acronym">BNF</orgName><desc><address><country key="FR"></country></address></desc></org></tutelle></tutelles></hal:affiliation></affiliation></author><author><name sortKey="Moreux, Jean Philippe" sort="Moreux, Jean Philippe" uniqKey="Moreux J" first="Jean-Philippe" last="Moreux">Jean-Philippe Moreux</name><affiliation><hal:affiliation type="laboratory" xml:id="struct-205116" status="VALID"><orgName>Bibliothèque nationale de France, Département de la Conservation</orgName><orgName type="acronym">BnF_DSC</orgName><desc><address><addrLine>Quai François Mauriac, 75706 Paris cedex 13</addrLine><country key="FR"></country></address><ref type="url">http://www.bnf.fr/fr/professionnels/conservation.html</ref></desc><listRelation><relation active="#struct-300057" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-300057" type="direct"><org type="institution" xml:id="struct-300057" status="VALID"><orgName>Bibliothèque Nationale de France</orgName><orgName type="acronym">BNF</orgName><desc><address><country key="FR"></country></address></desc></org></tutelle></tutelles></hal:affiliation></affiliation></author><author><name sortKey="Ragot, Nicolas" sort="Ragot, Nicolas" uniqKey="Ragot N" first="Nicolas" last="Ragot">Nicolas Ragot</name><affiliation><hal:affiliation type="laboratory" xml:id="struct-204893" status="VALID"><orgName>Laboratoire d'Informatique de l'Université de Tours</orgName><orgName type="acronym">LI</orgName><desc><address><addrLine>64, Avenue Jean Portalis, 37200 Tours</addrLine><country key="FR"></country></address><ref type="url">http://www.li.univ-tours.fr/</ref></desc><listRelation><relation active="#struct-300408" type="direct"></relation><relation name="EA6300" active="#struct-300298" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-300408" type="direct"><org type="institution" xml:id="struct-300408" status="VALID"><orgName>Polytech'Tours</orgName><desc><address><country key="FR"></country></address></desc></org></tutelle><tutelle name="EA6300" active="#struct-300298" type="direct"><org type="institution" xml:id="struct-300298" status="VALID"><orgName>Université François Rabelais - Tours</orgName><desc><address><addrLine>60 rue du Plat d'Étain, 37020 Tours cedex 1 </addrLine><country key="FR"></country></address><ref type="url">http://www.univ-tours.fr</ref></desc></org></tutelle></tutelles></hal:affiliation></affiliation></author><author><name sortKey="Paquet, Thierry" sort="Paquet, Thierry" uniqKey="Paquet T" first="Thierry" last="Paquet">Thierry Paquet</name><affiliation><hal:affiliation type="laboratory" xml:id="struct-23832" status="VALID"><orgName>Laboratoire d'Informatique, de Traitement de l'Information et des Systèmes</orgName><orgName type="acronym">LITIS</orgName><desc><address><addrLine>Avenue de l'Université UFR des Sciences et Techniques 76800 Saint-Etienne du Rouvray</addrLine><country key="FR"></country></address><ref type="url">http://www.litislab.eu</ref></desc><listRelation><relation active="#struct-300317" type="direct"></relation><relation name="EA4108" active="#struct-300318" type="direct"></relation><relation active="#struct-301288" type="direct"></relation><relation active="#struct-301232" type="indirect"></relation></listRelation><tutelles><tutelle active="#struct-300317" type="direct"><org type="institution" xml:id="struct-300317" status="VALID"><orgName>Université du Havre</orgName><desc><address><country key="FR"></country></address></desc></org></tutelle><tutelle name="EA4108" active="#struct-300318" type="direct"><org type="institution" xml:id="struct-300318" status="VALID"><orgName>Université de Rouen</orgName><desc><address><addrLine> 1 rue Thomas Becket - 76821 Mont-Saint-Aignan</addrLine><country key="FR"></country></address><ref type="url">http://www.univ-rouen.fr/</ref></desc></org></tutelle><tutelle active="#struct-301288" type="direct"><org type="department" xml:id="struct-301288" status="VALID"><orgName>Institut National des Sciences Appliquées - Rouen</orgName><orgName type="acronym">INSA Rouen</orgName><desc><address><country key="FR"></country></address></desc><listRelation><relation active="#struct-301232" type="direct"></relation></listRelation></org></tutelle><tutelle active="#struct-301232" type="indirect"><org type="institution" xml:id="struct-301232" status="VALID"><orgName>Institut National des Sciences Appliquées</orgName><orgName type="acronym">INSA</orgName><desc><address><country key="FR"></country></address></desc></org></tutelle></tutelles></hal:affiliation></affiliation></author></analytic></biblStruct></sourceDesc></fileDesc><profileDesc><textClass><keywords scheme="mix" xml:lang="en"><term>Historical documents digitization</term><term>Mass digitization projects</term><term>OCR quality assessment</term><term>Support Vector Regression (SVR)</term></keywords></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Since 2006 the national library of France (BnF) has developed many mass digitization projects on its collections. The indexation of digital documents on Gallica (the digital library of the BnF) is done through their textual content obtained thanks to service providers that use Optical Character Recognition software (OCR). The modern technologies of OCR achieve good performances on modern documents produced with uniform layout and known fonts. However, for old documents, OCR results are of lower quality. The OCR quality assessment is a real challenge for the BnF. On the one hand, due to the sequential architecture of OCR treatments, the identification of OCR errors sources is intractable. On the other hand, besides the word confidence, no additional quality information is reported in OCR outputs. In this paper, we present a study on OCR performance estimation aiming to control the quality of word transcriptions achieved by OCR. This quality assessment process has to operate without any comparison with ground truthed data. In this respect, our methodology relies on cross alignment of the OCR results with those of a secondary OCR called reference OCR. This secondary OCR provides uncertain but useful information that will be used as uncertain groundtruth. OCR performance is estimated using support vector regression. This predictor uses some global features computed on the cross-alignment results. The experimentations reported show that our estimate describes more faithfully the quality of OCR outputs than average word confidence scores that are computed by OCR. The proposed methodology can be adapted easily to various corpora by tuning the system using a training dataset of documents that have similar properties to those to be treated.</div></front></TEI><hal api="V3"><titleStmt><title xml:lang="en">OCR performance prediction using cross-OCR alignment</title><author role="aut"><persName><forename type="first">Ahmed</forename><surname>Ben Salah</surname></persName><email>ahmed.ben-salah@bnf.fr</email><ptr type="url" target="http://www.univ-rouen.fr/34392/0/fiche_annuaire/"></ptr><idno type="halauthor">768392</idno><affiliation ref="#struct-205116"></affiliation><affiliation ref="#struct-23832"></affiliation></author><author role="aut"><persName><forename type="first">Jean-Philippe</forename><surname>Moreux</surname></persName><email>jean-philippe.moreux@bnf.fr</email><idno type="halauthor">1102283</idno><affiliation ref="#struct-205116"></affiliation></author><author role="aut"><persName><forename type="first">Nicolas</forename><surname>Ragot</surname></persName><email>nicolas.ragot@univ-tours.fr</email><idno type="idhal">nicolas-ragot</idno><idno type="halauthor">847211</idno><affiliation ref="#struct-204893"></affiliation></author><author role="aut"><persName><forename type="first">Thierry</forename><surname>Paquet</surname></persName><email>thierry.paquet@univ-rouen.fr</email><idno type="halauthor">134993</idno><affiliation ref="#struct-23832"></affiliation></author><editor role="depositor"><persName><forename>Nicolas</forename><surname>Ragot</surname></persName><email>nicolas.ragot@univ-tours.fr</email></editor></titleStmt><editionStmt><edition n="v1" type="current"><date type="whenSubmitted">2015-09-02 13:33:57</date><date type="whenWritten">2015-02-16</date><date type="whenModified">2016-04-28 15:43:55</date><date type="whenReleased">2015-09-02 13:33:57</date><date type="whenProduced">2015-08-23</date></edition><respStmt><resp>contributor</resp><name key="119606"><persName><forename>Nicolas</forename><surname>Ragot</surname></persName><email>nicolas.ragot@univ-tours.fr</email></name></respStmt></editionStmt><publicationStmt><distributor>CCSD</distributor><idno type="halId">hal-01191701</idno><idno type="halUri">https://hal.archives-ouvertes.fr/hal-01191701</idno><idno type="halBibtex">bensalah:hal-01191701</idno><idno type="halRefHtml">13th International Conference on Document Analysis and Recognition (ICDAR 2015), Aug 2015, Nancy, France. 2015</idno><idno type="halRef">13th International Conference on Document Analysis and Recognition (ICDAR 2015), Aug 2015, Nancy, France. 2015</idno></publicationStmt><seriesStmt><idno type="stamp" n="UNIV-TOURS">Université François Rabelais</idno><idno type="stamp" n="UNIV-LEHAVRE">Université du Havre</idno><idno type="stamp" n="UNIV-ROUEN">Université de Rouen</idno><idno type="stamp" n="BNF">Bibliothèque nationale de France</idno><idno type="stamp" n="LITIS">Laboratoire d'Informatique, de Traitement de l'Information et des Systèmes</idno><idno type="stamp" n="COMUE-NORMANDIE">Normandie Université</idno><idno type="stamp" n="LIRFAI">Equipe RFAI du Laboratoire Informatique de Tours</idno><idno type="stamp" n="BNF_DSC" p="BNF">Bibliothèque nationale de France, Département de la Conservation</idno></seriesStmt><notesStmt><note type="audience" n="2">International</note><note type="invited" n="0">No</note><note type="popular" n="0">No</note><note type="peer" n="1">Yes</note><note type="proceedings" n="1">Yes</note></notesStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">OCR performance prediction using cross-OCR alignment</title><author role="aut"><persName><forename type="first">Ahmed</forename><surname>Ben Salah</surname></persName><email>ahmed.ben-salah@bnf.fr</email><ptr type="url" target="http://www.univ-rouen.fr/34392/0/fiche_annuaire/"></ptr><idno type="halAuthorId">768392</idno><affiliation ref="#struct-205116"></affiliation><affiliation ref="#struct-23832"></affiliation></author><author role="aut"><persName><forename type="first">Jean-Philippe</forename><surname>Moreux</surname></persName><email>jean-philippe.moreux@bnf.fr</email><idno type="halAuthorId">1102283</idno><affiliation ref="#struct-205116"></affiliation></author><author role="aut"><persName><forename type="first">Nicolas</forename><surname>Ragot</surname></persName><email>nicolas.ragot@univ-tours.fr</email><idno type="idHal">nicolas-ragot</idno><idno type="halAuthorId">847211</idno><affiliation ref="#struct-204893"></affiliation></author><author role="aut"><persName><forename type="first">Thierry</forename><surname>Paquet</surname></persName><email>thierry.paquet@univ-rouen.fr</email><idno type="halAuthorId">134993</idno><affiliation ref="#struct-23832"></affiliation></author></analytic><monogr><meeting><title>13th International Conference on Document Analysis and Recognition (ICDAR 2015)</title><date type="start">2015-08-23</date><settlement>Nancy</settlement><country key="FR">France</country></meeting><imprint><date type="datePub">2015-08-23</date></imprint></monogr></biblStruct></sourceDesc><profileDesc><langUsage><language ident="en">English</language></langUsage><textClass><keywords scheme="author"><term xml:lang="en">OCR quality assessment</term><term xml:lang="en">Historical documents digitization</term><term xml:lang="en">Support Vector Regression (SVR)</term><term xml:lang="en">Mass digitization projects</term></keywords><classCode scheme="halDomain" n="info.info-tt">Computer Science [cs]/Document and Text Processing</classCode><classCode scheme="halTypology" n="COMM">Conference papers</classCode></textClass><abstract xml:lang="en">Since 2006 the national library of France (BnF) has developed many mass digitization projects on its collections. The indexation of digital documents on Gallica (the digital library of the BnF) is done through their textual content obtained thanks to service providers that use Optical Character Recognition software (OCR). The modern technologies of OCR achieve good performances on modern documents produced with uniform layout and known fonts. However, for old documents, OCR results are of lower quality. The OCR quality assessment is a real challenge for the BnF. On the one hand, due to the sequential architecture of OCR treatments, the identification of OCR errors sources is intractable. On the other hand, besides the word confidence, no additional quality information is reported in OCR outputs. In this paper, we present a study on OCR performance estimation aiming to control the quality of word transcriptions achieved by OCR. This quality assessment process has to operate without any comparison with ground truthed data. In this respect, our methodology relies on cross alignment of the OCR results with those of a secondary OCR called reference OCR. This secondary OCR provides uncertain but useful information that will be used as uncertain groundtruth. OCR performance is estimated using support vector regression. This predictor uses some global features computed on the cross-alignment results. The experimentations reported show that our estimate describes more faithfully the quality of OCR outputs than average word confidence scores that are computed by OCR. The proposed methodology can be adapted easily to various corpora by tuning the system using a training dataset of documents that have similar properties to those to be treated.</abstract></profileDesc></hal></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Hal/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000089 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Hal/Corpus/biblio.hfd -nk 000089 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Ticri/CIDE
|area= OcrV1
|flux= Hal
|étape= Corpus
|type= RBID
|clé= Hal:hal-01191701
|texte= OCR performance prediction using cross-OCR alignment
}}
| This area was generated with Dilib version V0.6.32. |  |