Applying SIMD to Optical Character Recognition (OCR)
Identifieur interne : 000D28 ( Main/Exploration ); précédent : 000D27; suivant : 000D29Applying SIMD to Optical Character Recognition (OCR)
Auteurs : GUAN YU [Belgique] ; Lafruit Gauthier [Belgique] ; Richard Stahl [Belgique] ; Henk Corporaal [Pays-Bas] ; Peter Schelkens [Belgique]Source :
- Proceedings of SPIE - The International Society for Optical Engineering [ 0277-786X ] ; 2008.
Descripteurs français
- Pascal (Inist)
- Wicri :
- topic : Automatisation.
English descriptors
- KwdEn :
Abstract
Optical Character Recognition (OCR) techniques are widely used in data/text entry, process automation. Decades of research efforts have made the accurate recognition of typewritten text largely accepted as a solved problem. Driven by practical usage demands, the low complexity and high performance implementation techniques of OCR systems are studied. Recent research shows that it may not be possible even for a simple OCR to run on a portable device without a specialized digital signal processor. In this paper, we present a highly data-parallelized implementation of OCR for typewritten text onto the linear processor array of the Xetal chip. Besides the preprocessing stage, the most computation intensive part of OCR recognizing individual characters is highly parallelized onto the Single Instruction Multiple Data (SIMD) engine of the Xetal chip, which can process a VGA-resolution text frame within one tenth of a second. In addition, different parallelization schemes are explored to make trade-off between the degree of parallelism and the costs of preprocessing to reorganize data to feed the SIMD engine and post-processing to assemble and collect results. The exploration of parallelized OCR application brings additional performance gain when mapped onto the linear processor array of the Xetal chip.
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream PascalFrancis, to step Corpus: 000244
- to stream PascalFrancis, to step Curation: 000535
- to stream PascalFrancis, to step Checkpoint: 000250
- to stream Main, to step Merge: 000D40
- to stream Main, to step Curation: 000D28
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">Applying SIMD to Optical Character Recognition (OCR)</title><author><name sortKey="Guan Yu" sort="Guan Yu" uniqKey="Guan Yu" last="Guan Yu">GUAN YU</name><affiliation wicri:level="3"><inist:fA14 i1="01"><s1>Dept. of Electronics and Information Processing, Vrije Universiteit Brussel</s1><s2>Brussels</s2><s3>BEL</s3><sZ>1 aut.</sZ><sZ>5 aut.</sZ></inist:fA14><country>Belgique</country><placeName><settlement type="city">Bruxelles</settlement><region nuts="2">Région de Bruxelles-Capitale</region></placeName></affiliation><affiliation wicri:level="1"><inist:fA14 i1="02"><s1>Nomadic Embedded Systems, Inter-university Micro-Electronics Centre</s1><s2>Leuven</s2><s3>BEL</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ><sZ>3 aut.</sZ></inist:fA14><country>Belgique</country><wicri:noRegion>Leuven</wicri:noRegion></affiliation></author><author><name sortKey="Gauthier, Lafruit" sort="Gauthier, Lafruit" uniqKey="Gauthier L" first="Lafruit" last="Gauthier">Lafruit Gauthier</name><affiliation wicri:level="1"><inist:fA14 i1="02"><s1>Nomadic Embedded Systems, Inter-university Micro-Electronics Centre</s1><s2>Leuven</s2><s3>BEL</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ><sZ>3 aut.</sZ></inist:fA14><country>Belgique</country><wicri:noRegion>Leuven</wicri:noRegion></affiliation></author><author><name sortKey="Stahl, Richard" sort="Stahl, Richard" uniqKey="Stahl R" first="Richard" last="Stahl">Richard Stahl</name><affiliation wicri:level="1"><inist:fA14 i1="02"><s1>Nomadic Embedded Systems, Inter-university Micro-Electronics Centre</s1><s2>Leuven</s2><s3>BEL</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ><sZ>3 aut.</sZ></inist:fA14><country>Belgique</country><wicri:noRegion>Leuven</wicri:noRegion></affiliation></author><author><name sortKey="Corporaal, Henk" sort="Corporaal, Henk" uniqKey="Corporaal H" first="Henk" last="Corporaal">Henk Corporaal</name><affiliation wicri:level="1"><inist:fA14 i1="03"><s1>Faculty of Electrical Engineering, Technical University of Eindhoven</s1><s2>Eindhoven</s2><s3>NLD</s3><sZ>4 aut.</sZ></inist:fA14><country>Pays-Bas</country><wicri:noRegion>Eindhoven</wicri:noRegion></affiliation></author><author><name sortKey="Schelkens, Peter" sort="Schelkens, Peter" uniqKey="Schelkens P" first="Peter" last="Schelkens">Peter Schelkens</name><affiliation wicri:level="3"><inist:fA14 i1="01"><s1>Dept. of Electronics and Information Processing, Vrije Universiteit Brussel</s1><s2>Brussels</s2><s3>BEL</s3><sZ>1 aut.</sZ><sZ>5 aut.</sZ></inist:fA14><country>Belgique</country><placeName><settlement type="city">Bruxelles</settlement><region nuts="2">Région de Bruxelles-Capitale</region></placeName></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">INIST</idno><idno type="inist">09-0009802</idno><date when="2008">2008</date><idno type="stanalyst">PASCAL 09-0009802 INIST</idno><idno type="RBID">Pascal:09-0009802</idno><idno type="wicri:Area/PascalFrancis/Corpus">000244</idno><idno type="wicri:Area/PascalFrancis/Curation">000535</idno><idno type="wicri:Area/PascalFrancis/Checkpoint">000250</idno><idno type="wicri:doubleKey">0277-786X:2008:Guan Yu:applying:simd:to</idno><idno type="wicri:Area/Main/Merge">000D40</idno><idno type="wicri:Area/Main/Curation">000D28</idno><idno type="wicri:Area/Main/Exploration">000D28</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">Applying SIMD to Optical Character Recognition (OCR)</title><author><name sortKey="Guan Yu" sort="Guan Yu" uniqKey="Guan Yu" last="Guan Yu">GUAN YU</name><affiliation wicri:level="3"><inist:fA14 i1="01"><s1>Dept. of Electronics and Information Processing, Vrije Universiteit Brussel</s1><s2>Brussels</s2><s3>BEL</s3><sZ>1 aut.</sZ><sZ>5 aut.</sZ></inist:fA14><country>Belgique</country><placeName><settlement type="city">Bruxelles</settlement><region nuts="2">Région de Bruxelles-Capitale</region></placeName></affiliation><affiliation wicri:level="1"><inist:fA14 i1="02"><s1>Nomadic Embedded Systems, Inter-university Micro-Electronics Centre</s1><s2>Leuven</s2><s3>BEL</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ><sZ>3 aut.</sZ></inist:fA14><country>Belgique</country><wicri:noRegion>Leuven</wicri:noRegion></affiliation></author><author><name sortKey="Gauthier, Lafruit" sort="Gauthier, Lafruit" uniqKey="Gauthier L" first="Lafruit" last="Gauthier">Lafruit Gauthier</name><affiliation wicri:level="1"><inist:fA14 i1="02"><s1>Nomadic Embedded Systems, Inter-university Micro-Electronics Centre</s1><s2>Leuven</s2><s3>BEL</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ><sZ>3 aut.</sZ></inist:fA14><country>Belgique</country><wicri:noRegion>Leuven</wicri:noRegion></affiliation></author><author><name sortKey="Stahl, Richard" sort="Stahl, Richard" uniqKey="Stahl R" first="Richard" last="Stahl">Richard Stahl</name><affiliation wicri:level="1"><inist:fA14 i1="02"><s1>Nomadic Embedded Systems, Inter-university Micro-Electronics Centre</s1><s2>Leuven</s2><s3>BEL</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ><sZ>3 aut.</sZ></inist:fA14><country>Belgique</country><wicri:noRegion>Leuven</wicri:noRegion></affiliation></author><author><name sortKey="Corporaal, Henk" sort="Corporaal, Henk" uniqKey="Corporaal H" first="Henk" last="Corporaal">Henk Corporaal</name><affiliation wicri:level="1"><inist:fA14 i1="03"><s1>Faculty of Electrical Engineering, Technical University of Eindhoven</s1><s2>Eindhoven</s2><s3>NLD</s3><sZ>4 aut.</sZ></inist:fA14><country>Pays-Bas</country><wicri:noRegion>Eindhoven</wicri:noRegion></affiliation></author><author><name sortKey="Schelkens, Peter" sort="Schelkens, Peter" uniqKey="Schelkens P" first="Peter" last="Schelkens">Peter Schelkens</name><affiliation wicri:level="3"><inist:fA14 i1="01"><s1>Dept. of Electronics and Information Processing, Vrije Universiteit Brussel</s1><s2>Brussels</s2><s3>BEL</s3><sZ>1 aut.</sZ><sZ>5 aut.</sZ></inist:fA14><country>Belgique</country><placeName><settlement type="city">Bruxelles</settlement><region nuts="2">Région de Bruxelles-Capitale</region></placeName></affiliation></author></analytic><series><title level="j" type="main">Proceedings of SPIE - The International Society for Optical Engineering</title><idno type="ISSN">0277-786X</idno><imprint><date when="2008">2008</date></imprint></series></biblStruct></sourceDesc><seriesStmt><title level="j" type="main">Proceedings of SPIE - The International Society for Optical Engineering</title><idno type="ISSN">0277-786X</idno></seriesStmt></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Automation</term><term>Image processing</term></keywords><keywords scheme="Pascal" xml:lang="fr"><term>Automatisation</term><term>Traitement image</term><term>0130C</term><term>0705P</term><term>4230V</term></keywords><keywords scheme="Wicri" type="topic" xml:lang="fr"><term>Automatisation</term></keywords></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Optical Character Recognition (OCR) techniques are widely used in data/text entry, process automation. Decades of research efforts have made the accurate recognition of typewritten text largely accepted as a solved problem. Driven by practical usage demands, the low complexity and high performance implementation techniques of OCR systems are studied. Recent research shows that it may not be possible even for a simple OCR to run on a portable device without a specialized digital signal processor. In this paper, we present a highly data-parallelized implementation of OCR for typewritten text onto the linear processor array of the Xetal chip. Besides the preprocessing stage, the most computation intensive part of OCR recognizing individual characters is highly parallelized onto the Single Instruction Multiple Data (SIMD) engine of the Xetal chip, which can process a VGA-resolution text frame within one tenth of a second. In addition, different parallelization schemes are explored to make trade-off between the degree of parallelism and the costs of preprocessing to reorganize data to feed the SIMD engine and post-processing to assemble and collect results. The exploration of parallelized OCR application brings additional performance gain when mapped onto the linear processor array of the Xetal chip.</div></front></TEI><affiliations><list><country><li>Belgique</li><li>Pays-Bas</li></country><region><li>Région de Bruxelles-Capitale</li></region><settlement><li>Bruxelles</li></settlement></list><tree><country name="Belgique"><region name="Région de Bruxelles-Capitale"><name sortKey="Guan Yu" sort="Guan Yu" uniqKey="Guan Yu" last="Guan Yu">GUAN YU</name></region><name sortKey="Gauthier, Lafruit" sort="Gauthier, Lafruit" uniqKey="Gauthier L" first="Lafruit" last="Gauthier">Lafruit Gauthier</name><name sortKey="Guan Yu" sort="Guan Yu" uniqKey="Guan Yu" last="Guan Yu">GUAN YU</name><name sortKey="Schelkens, Peter" sort="Schelkens, Peter" uniqKey="Schelkens P" first="Peter" last="Schelkens">Peter Schelkens</name><name sortKey="Stahl, Richard" sort="Stahl, Richard" uniqKey="Stahl R" first="Richard" last="Stahl">Richard Stahl</name></country><country name="Pays-Bas"><noRegion><name sortKey="Corporaal, Henk" sort="Corporaal, Henk" uniqKey="Corporaal H" first="Henk" last="Corporaal">Henk Corporaal</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 000D28 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000D28 | 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:09-0009802
|texte= Applying SIMD to Optical Character Recognition (OCR)
}}
| This area was generated with Dilib version V0.6.32. |  |