Multiple classifier for degraded machine printed character recognition
Identifieur interne : 000B03 ( Main/Exploration ); précédent : 000B02; suivant : 000B04Multiple classifier for degraded machine printed character recognition
Auteurs : A. Namane [Algérie] ; P. Meyrueis [France]Source :
English descriptors
- mix :
Abstract
The general problem of optical character recognition (OCR) remains a fundamental but not entirely solved issue in document analysis. In spite of significant improvements in the area of optical character recognition, the recognition of degraded printed characters, in particular, is still lacking satisfactory solutions. This paper presents an OCR method that combines the Hopfield network with a set of autoassociators for degraded character recognition. In the serial combination, the first classifier must achieve lower errors and be very well suited for rejection, whereas the second classifier must allow only low errors and rejects. A relative distance is used as a quality measurement parameter which makes the Hopfieldbased classifier very powerful and very well suited for rejection. We report experimental results for a comparison of three methods: the Hopfield model, the autoassociator-based classifier and the proposed combined architecture. Experimental results show the ability of the model to yield relevant and robust recognition with no errors on poor quality bank check characters even when the patterns are highly degraded.
Url:
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream Hal, to step Corpus: 000085
- to stream Hal, to step Curation: 000085
- to stream Hal, to step Checkpoint: 000119
- to stream Main, to step Merge: 000B14
- to stream Main, to step Curation: 000B03
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Multiple classifier for degraded machine printed character recognition</title><author><name sortKey="Namane, A" sort="Namane, A" uniqKey="Namane A" first="A." last="Namane">A. Namane</name><affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-81752" status="VALID"><orgName>Laboratoire de Traitement du Signal et de l'Image</orgName><orgName type="acronym">LTSI</orgName><desc><address><addrLine>Université de Saâd Dahleb de Blida, Faculté des Sciences de l'Ingénieur, département d'Electronique</addrLine><country key="DZ"></country></address></desc><listRelation><relation active="#struct-320871" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-320871" type="direct"><org type="institution" xml:id="struct-320871" status="INCOMING"><orgName>Université de Saâd Dahleb de Blida, Faculté des Sciences de l'Ingénieur, département d'Electronique,</orgName><desc><address><country key="DZ"></country></address></desc></org></tutelle></tutelles></hal:affiliation><country>Algérie</country></affiliation></author><author><name sortKey="Meyrueis, P" sort="Meyrueis, P" uniqKey="Meyrueis P" first="P." last="Meyrueis">P. Meyrueis</name><affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-155219" status="VALID"><orgName>Laboratoire des Systèmes Photoniques</orgName><orgName type="acronym">LSP</orgName><desc><address><addrLine>Boulevard Sébastien Brant 67412 Strasbourg</addrLine><country key="FR"></country></address><ref type="url">http://www.unistra.fr/index.php?id=1828</ref></desc><listRelation><relation active="#struct-302445" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-302445" type="direct"><org type="institution" xml:id="struct-302445" status="VALID"><orgName>Université de Strasbourg</orgName><desc><address><country key="FR"></country></address></desc></org></tutelle></tutelles></hal:affiliation><country>France</country><placeName><settlement type="city">Strasbourg</settlement><region type="region" nuts="2">Alsace</region></placeName><orgName type="university">Université de Strasbourg</orgName></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">HAL</idno><idno type="RBID">Hal:hal-00334417</idno><idno type="halId">hal-00334417</idno><idno type="halUri">https://hal.archives-ouvertes.fr/hal-00334417</idno><idno type="url">https://hal.archives-ouvertes.fr/hal-00334417</idno><date when="2008-10">2008-10</date><idno type="wicri:Area/Hal/Corpus">000085</idno><idno type="wicri:Area/Hal/Curation">000085</idno><idno type="wicri:Area/Hal/Checkpoint">000119</idno><idno type="wicri:Area/Main/Merge">000B14</idno><idno type="wicri:Area/Main/Curation">000B03</idno><idno type="wicri:Area/Main/Exploration">000B03</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">Multiple classifier for degraded machine printed character recognition</title><author><name sortKey="Namane, A" sort="Namane, A" uniqKey="Namane A" first="A." last="Namane">A. Namane</name><affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-81752" status="VALID"><orgName>Laboratoire de Traitement du Signal et de l'Image</orgName><orgName type="acronym">LTSI</orgName><desc><address><addrLine>Université de Saâd Dahleb de Blida, Faculté des Sciences de l'Ingénieur, département d'Electronique</addrLine><country key="DZ"></country></address></desc><listRelation><relation active="#struct-320871" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-320871" type="direct"><org type="institution" xml:id="struct-320871" status="INCOMING"><orgName>Université de Saâd Dahleb de Blida, Faculté des Sciences de l'Ingénieur, département d'Electronique,</orgName><desc><address><country key="DZ"></country></address></desc></org></tutelle></tutelles></hal:affiliation><country>Algérie</country></affiliation></author><author><name sortKey="Meyrueis, P" sort="Meyrueis, P" uniqKey="Meyrueis P" first="P." last="Meyrueis">P. Meyrueis</name><affiliation wicri:level="1"><hal:affiliation type="laboratory" xml:id="struct-155219" status="VALID"><orgName>Laboratoire des Systèmes Photoniques</orgName><orgName type="acronym">LSP</orgName><desc><address><addrLine>Boulevard Sébastien Brant 67412 Strasbourg</addrLine><country key="FR"></country></address><ref type="url">http://www.unistra.fr/index.php?id=1828</ref></desc><listRelation><relation active="#struct-302445" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-302445" type="direct"><org type="institution" xml:id="struct-302445" status="VALID"><orgName>Université de Strasbourg</orgName><desc><address><country key="FR"></country></address></desc></org></tutelle></tutelles></hal:affiliation><country>France</country><placeName><settlement type="city">Strasbourg</settlement><region type="region" nuts="2">Alsace</region></placeName><orgName type="university">Université de Strasbourg</orgName></affiliation></author></analytic></biblStruct></sourceDesc></fileDesc><profileDesc><textClass><keywords scheme="mix" xml:lang="en"><term>Hopfield model</term><term>OCR</term><term>associative memory</term><term>autoassociator</term><term>character recognition</term><term>character recognition.</term><term>degraded printed characters</term><term>serial combination</term></keywords></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">The general problem of optical character recognition (OCR) remains a fundamental but not entirely solved issue in document analysis. In spite of significant improvements in the area of optical character recognition, the recognition of degraded printed characters, in particular, is still lacking satisfactory solutions. This paper presents an OCR method that combines the Hopfield network with a set of autoassociators for degraded character recognition. In the serial combination, the first classifier must achieve lower errors and be very well suited for rejection, whereas the second classifier must allow only low errors and rejects. A relative distance is used as a quality measurement parameter which makes the Hopfieldbased classifier very powerful and very well suited for rejection. We report experimental results for a comparison of three methods: the Hopfield model, the autoassociator-based classifier and the proposed combined architecture. Experimental results show the ability of the model to yield relevant and robust recognition with no errors on poor quality bank check characters even when the patterns are highly degraded.</div></front></TEI><affiliations><list><country><li>Algérie</li><li>France</li></country><region><li>Alsace</li></region><settlement><li>Strasbourg</li></settlement><orgName><li>Université de Strasbourg</li></orgName></list><tree><country name="Algérie"><noRegion><name sortKey="Namane, A" sort="Namane, A" uniqKey="Namane A" first="A." last="Namane">A. Namane</name></noRegion></country><country name="France"><region name="Alsace"><name sortKey="Meyrueis, P" sort="Meyrueis, P" uniqKey="Meyrueis P" first="P." last="Meyrueis">P. Meyrueis</name></region></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 000B03 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000B03 | 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-00334417
|texte= Multiple classifier for degraded machine printed character recognition
}}
| This area was generated with Dilib version V0.6.32. |  |