Using consensus sequence voting to correct OCR errors
Identifieur interne : 002699 ( Main/Merge ); précédent : 002698; suivant : 002700Using consensus sequence voting to correct OCR errors
Auteurs : D. Lopresti [États-Unis] ; J. Zhou [États-Unis]Source :
- Computer vision and image understanding [ 1077-3142 ] ; 1997.
Descripteurs français
- Pascal (Inist)
- Wicri :
- topic : Vote.
English descriptors
- KwdEn :
Abstract
We present experimental results suggesting that between 20 and 50% of the errors caused by a single OCR package can be eliminated by simply scanning a page three times and running a "consensus sequence" voting procedure. This technique, which originates from molecular biology, takes exponential time in general, but can be specialized to a fast heuristic guaranteed to be optimal for the cases of interest. The improvement in recognition accuracy is achieved without making a priori assumptions about the distribution of OCR errors (i.e., no "training" is required).
Links toward previous steps (curation, corpus...)
- to stream PascalFrancis, to step Corpus: 000935
- to stream PascalFrancis, to step Curation: 000A64
- to stream PascalFrancis, to step Checkpoint: 000868
Links to Exploration step
Pascal:97-0362902Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">Using consensus sequence voting to correct OCR errors</title><author><name sortKey="Lopresti, D" sort="Lopresti, D" uniqKey="Lopresti D" first="D." last="Lopresti">D. Lopresti</name><affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Matsushita Information Technology Laboratory, Panasonic Technologies, Inc., Two Research Way</s1><s2>Princeton, New Jersey 08540</s2><s3>USA</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ></inist:fA14><country>États-Unis</country><wicri:noRegion>Princeton, New Jersey 08540</wicri:noRegion></affiliation></author><author><name sortKey="Zhou, J" sort="Zhou, J" uniqKey="Zhou J" first="J." last="Zhou">J. Zhou</name><affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Matsushita Information Technology Laboratory, Panasonic Technologies, Inc., Two Research Way</s1><s2>Princeton, New Jersey 08540</s2><s3>USA</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ></inist:fA14><country>États-Unis</country><wicri:noRegion>Princeton, New Jersey 08540</wicri:noRegion></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">INIST</idno><idno type="inist">97-0362902</idno><date when="1997">1997</date><idno type="stanalyst">PASCAL 97-0362902 INIST</idno><idno type="RBID">Pascal:97-0362902</idno><idno type="wicri:Area/PascalFrancis/Corpus">000935</idno><idno type="wicri:Area/PascalFrancis/Curation">000A64</idno><idno type="wicri:Area/PascalFrancis/Checkpoint">000868</idno><idno type="wicri:doubleKey">1077-3142:1997:Lopresti D:using:consensus:sequence</idno><idno type="wicri:Area/Main/Merge">002699</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">Using consensus sequence voting to correct OCR errors</title><author><name sortKey="Lopresti, D" sort="Lopresti, D" uniqKey="Lopresti D" first="D." last="Lopresti">D. Lopresti</name><affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Matsushita Information Technology Laboratory, Panasonic Technologies, Inc., Two Research Way</s1><s2>Princeton, New Jersey 08540</s2><s3>USA</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ></inist:fA14><country>États-Unis</country><wicri:noRegion>Princeton, New Jersey 08540</wicri:noRegion></affiliation></author><author><name sortKey="Zhou, J" sort="Zhou, J" uniqKey="Zhou J" first="J." last="Zhou">J. Zhou</name><affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Matsushita Information Technology Laboratory, Panasonic Technologies, Inc., Two Research Way</s1><s2>Princeton, New Jersey 08540</s2><s3>USA</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ></inist:fA14><country>États-Unis</country><wicri:noRegion>Princeton, New Jersey 08540</wicri:noRegion></affiliation></author></analytic><series><title level="j" type="main">Computer vision and image understanding</title><title level="j" type="abbreviated">Comput. vis. image underst.</title><idno type="ISSN">1077-3142</idno><imprint><date when="1997">1997</date></imprint></series></biblStruct></sourceDesc><seriesStmt><title level="j" type="main">Computer vision and image understanding</title><title level="j" type="abbreviated">Comput. vis. image underst.</title><idno type="ISSN">1077-3142</idno></seriesStmt></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Algorithm</term><term>Character recognition</term><term>Error analysis</term><term>Experimental study</term><term>Image processing</term><term>System architecture</term><term>System performance</term><term>Voting</term></keywords><keywords scheme="Pascal" xml:lang="fr"><term>Traitement image</term><term>Reconnaissance caractère</term><term>Architecture système</term><term>Algorithme</term><term>Vote</term><term>Calcul erreur</term><term>Etude expérimentale</term><term>Performance système</term></keywords><keywords scheme="Wicri" type="topic" xml:lang="fr"><term>Vote</term></keywords></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">We present experimental results suggesting that between 20 and 50% of the errors caused by a single OCR package can be eliminated by simply scanning a page three times and running a "consensus sequence" voting procedure. This technique, which originates from molecular biology, takes exponential time in general, but can be specialized to a fast heuristic guaranteed to be optimal for the cases of interest. The improvement in recognition accuracy is achieved without making a priori assumptions about the distribution of OCR errors (i.e., no "training" is required).</div></front></TEI><affiliations><list><country><li>États-Unis</li></country></list><tree><country name="États-Unis"><noRegion><name sortKey="Lopresti, D" sort="Lopresti, D" uniqKey="Lopresti D" first="D." last="Lopresti">D. Lopresti</name></noRegion><name sortKey="Zhou, J" sort="Zhou, J" uniqKey="Zhou J" first="J." last="Zhou">J. Zhou</name></country></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Main/Merge
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 002699 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Merge/biblio.hfd -nk 002699 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Ticri/CIDE
|area= OcrV1
|flux= Main
|étape= Merge
|type= RBID
|clé= Pascal:97-0362902
|texte= Using consensus sequence voting to correct OCR errors
}}
| This area was generated with Dilib version V0.6.32. |  |