Performance Evaluation and Benchmarking of Six-Page Segmentation Algorithms
Identifieur interne : 000D07 ( Main/Exploration ); précédent : 000D06; suivant : 000D08Performance Evaluation and Benchmarking of Six-Page Segmentation Algorithms
Auteurs : Faisal Shafait [Allemagne] ; Daniel Keysers [Allemagne] ; Thomas M. Breuel [Allemagne]Source :
- IEEE transactions on pattern analysis and machine intelligence [ 0162-8828 ] ; 2008.
Descripteurs français
- Pascal (Inist)
- Wicri :
- topic : Intelligence artificielle, Base de données.
English descriptors
- KwdEn :
- Algorithms, Artificial Intelligence, Artificial intelligence, Automatic Data Processing (methods), Benchmarking, Character recognition, Computer Graphics, Database, Document processing, Documentation (methods), Ground truth, Image Enhancement (methods), Image Interpretation, Computer-Assisted (methods), Information Storage and Retrieval (methods), Metric, Models, Statistical, Numerical Analysis, Computer-Assisted, Optical character recognition, Optimization, Pattern Recognition, Automated (methods), Pattern analysis, Performance evaluation, Reproducibility of Results, Segmentation, Sensitivity and Specificity, Signal Processing, Computer-Assisted, Subtraction Technique, Text analysis, User-Computer Interface, Voronoï diagram.
- MESH :
- methods : Automatic Data Processing, Documentation, Image Enhancement, Image Interpretation, Computer-Assisted, Information Storage and Retrieval, Pattern Recognition, Automated.
- Algorithms, Artificial Intelligence, Benchmarking, Computer Graphics, Models, Statistical, Numerical Analysis, Computer-Assisted, Reproducibility of Results, Sensitivity and Specificity, Signal Processing, Computer-Assisted, Subtraction Technique, User-Computer Interface.
Abstract
-Informative benchmarks are crucial for optimizing the page segmentation step of an OCR system, frequently the performance limiting step for overall OCR system performance. We show that current evaluation scores are insufficient for diagnosing specific errors in page segmentation and fail to identify some classes of serious segmentation errors altogether. This paper introduces a vectorial score that is sensitive to, and identifies, the most important classes of segmentation errors (over, under, and mis-segmentation) and what page components (lines, blocks, etc.) are affected. Unlike previous schemes, our evaluation method has a canonical representation of ground-truth data and guarantees pixel-accurate evaluation results for arbitrary region shapes. We present the results of evaluating widely used segmentation algorithms (x-y cut, smearing, whitespace analysis, constrained text-line finding, docstrum, and Voronoi) on the UW-III database and demonstrate that the new evaluation scheme permits the identification of several specific flaws in individual segmentation methods.
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream PascalFrancis, to step Corpus: 000282
- to stream PascalFrancis, to step Curation: 000502
- to stream PascalFrancis, to step Checkpoint: 000229
- to stream Main, to step Merge: 000D19
- to stream PubMed, to step Corpus: 000050
- to stream PubMed, to step Curation: 000050
- to stream PubMed, to step Checkpoint: 000050
- to stream Ncbi, to step Merge: 000052
- to stream Ncbi, to step Curation: 000052
- to stream Ncbi, to step Checkpoint: 000052
- to stream Main, to step Merge: 000B24
- to stream Main, to step Curation: 000D07
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">Performance Evaluation and Benchmarking of Six-Page Segmentation Algorithms</title>
<author><name sortKey="Shafait, Faisal" sort="Shafait, Faisal" uniqKey="Shafait F" first="Faisal" last="Shafait">Faisal Shafait</name>
<affiliation wicri:level="3"><inist:fA14 i1="01"><s1>Image Understanding and Pattern Recognition Research Group, German Research Center for Artificial Intelligence (DFKI GmbH)</s1>
<s2>67663 Kaiserslautern</s2>
<s3>DEU</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>Allemagne</country>
<placeName><region type="land" nuts="2">Rhénanie-Palatinat</region>
<settlement type="city">Kaiserslautern</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Keysers, Daniel" sort="Keysers, Daniel" uniqKey="Keysers D" first="Daniel" last="Keysers">Daniel Keysers</name>
<affiliation wicri:level="3"><inist:fA14 i1="01"><s1>Image Understanding and Pattern Recognition Research Group, German Research Center for Artificial Intelligence (DFKI GmbH)</s1>
<s2>67663 Kaiserslautern</s2>
<s3>DEU</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>Allemagne</country>
<placeName><region type="land" nuts="2">Rhénanie-Palatinat</region>
<settlement type="city">Kaiserslautern</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Breuel, Thomas M" sort="Breuel, Thomas M" uniqKey="Breuel T" first="Thomas M." last="Breuel">Thomas M. Breuel</name>
<affiliation wicri:level="4"><inist:fA14 i1="02"><s1>Department of Computer Science, Technical University of Kaiserslautern</s1>
<s2>67663 Kaiserslautern</s2>
<s3>DEU</s3>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Allemagne</country>
<placeName><region type="land" nuts="2">Rhénanie-Palatinat</region>
<settlement type="city">Kaiserslautern</settlement>
</placeName>
<orgName type="university">Université de technologie de Kaiserslautern</orgName>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">INIST</idno>
<idno type="inist">08-0254155</idno>
<date when="2008">2008</date>
<idno type="stanalyst">PASCAL 08-0254155 INIST</idno>
<idno type="RBID">Pascal:08-0254155</idno>
<idno type="wicri:Area/PascalFrancis/Corpus">000282</idno>
<idno type="wicri:Area/PascalFrancis/Curation">000502</idno>
<idno type="wicri:Area/PascalFrancis/Checkpoint">000229</idno>
<idno type="wicri:doubleKey">0162-8828:2008:Shafait F:performance:evaluation:and</idno>
<idno type="wicri:Area/Main/Merge">000D19</idno>
<idno type="wicri:source">PubMed</idno>
<idno type="RBID">pubmed:18421102</idno>
<idno type="wicri:Area/PubMed/Corpus">000050</idno>
<idno type="wicri:Area/PubMed/Curation">000050</idno>
<idno type="wicri:Area/PubMed/Checkpoint">000050</idno>
<idno type="wicri:Area/Ncbi/Merge">000052</idno>
<idno type="wicri:Area/Ncbi/Curation">000052</idno>
<idno type="wicri:Area/Ncbi/Checkpoint">000052</idno>
<idno type="wicri:doubleKey">0162-8828:2008:Shafait F:performance:evaluation:and</idno>
<idno type="wicri:Area/Main/Merge">000B24</idno>
<idno type="wicri:Area/Main/Curation">000D07</idno>
<idno type="wicri:Area/Main/Exploration">000D07</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">Performance Evaluation and Benchmarking of Six-Page Segmentation Algorithms</title>
<author><name sortKey="Shafait, Faisal" sort="Shafait, Faisal" uniqKey="Shafait F" first="Faisal" last="Shafait">Faisal Shafait</name>
<affiliation wicri:level="3"><inist:fA14 i1="01"><s1>Image Understanding and Pattern Recognition Research Group, German Research Center for Artificial Intelligence (DFKI GmbH)</s1>
<s2>67663 Kaiserslautern</s2>
<s3>DEU</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>Allemagne</country>
<placeName><region type="land" nuts="2">Rhénanie-Palatinat</region>
<settlement type="city">Kaiserslautern</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Keysers, Daniel" sort="Keysers, Daniel" uniqKey="Keysers D" first="Daniel" last="Keysers">Daniel Keysers</name>
<affiliation wicri:level="3"><inist:fA14 i1="01"><s1>Image Understanding and Pattern Recognition Research Group, German Research Center for Artificial Intelligence (DFKI GmbH)</s1>
<s2>67663 Kaiserslautern</s2>
<s3>DEU</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>Allemagne</country>
<placeName><region type="land" nuts="2">Rhénanie-Palatinat</region>
<settlement type="city">Kaiserslautern</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Breuel, Thomas M" sort="Breuel, Thomas M" uniqKey="Breuel T" first="Thomas M." last="Breuel">Thomas M. Breuel</name>
<affiliation wicri:level="4"><inist:fA14 i1="02"><s1>Department of Computer Science, Technical University of Kaiserslautern</s1>
<s2>67663 Kaiserslautern</s2>
<s3>DEU</s3>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Allemagne</country>
<placeName><region type="land" nuts="2">Rhénanie-Palatinat</region>
<settlement type="city">Kaiserslautern</settlement>
</placeName>
<orgName type="university">Université de technologie de Kaiserslautern</orgName>
</affiliation>
</author>
</analytic>
<series><title level="j" type="main">IEEE transactions on pattern analysis and machine intelligence</title>
<title level="j" type="abbreviated">IEEE trans. pattern anal. mach. intell.</title>
<idno type="ISSN">0162-8828</idno>
<imprint><date when="2008">2008</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><title level="j" type="main">IEEE transactions on pattern analysis and machine intelligence</title>
<title level="j" type="abbreviated">IEEE trans. pattern anal. mach. intell.</title>
<idno type="ISSN">0162-8828</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Algorithms</term>
<term>Artificial Intelligence</term>
<term>Artificial intelligence</term>
<term>Automatic Data Processing (methods)</term>
<term>Benchmarking</term>
<term>Character recognition</term>
<term>Computer Graphics</term>
<term>Database</term>
<term>Document processing</term>
<term>Documentation (methods)</term>
<term>Ground truth</term>
<term>Image Enhancement (methods)</term>
<term>Image Interpretation, Computer-Assisted (methods)</term>
<term>Information Storage and Retrieval (methods)</term>
<term>Metric</term>
<term>Models, Statistical</term>
<term>Numerical Analysis, Computer-Assisted</term>
<term>Optical character recognition</term>
<term>Optimization</term>
<term>Pattern Recognition, Automated (methods)</term>
<term>Pattern analysis</term>
<term>Performance evaluation</term>
<term>Reproducibility of Results</term>
<term>Segmentation</term>
<term>Sensitivity and Specificity</term>
<term>Signal Processing, Computer-Assisted</term>
<term>Subtraction Technique</term>
<term>Text analysis</term>
<term>User-Computer Interface</term>
<term>Voronoï diagram</term>
</keywords>
<keywords scheme="MESH" qualifier="methods" xml:lang="en"><term>Automatic Data Processing</term>
<term>Documentation</term>
<term>Image Enhancement</term>
<term>Image Interpretation, Computer-Assisted</term>
<term>Information Storage and Retrieval</term>
<term>Pattern Recognition, Automated</term>
</keywords>
<keywords scheme="MESH" xml:lang="en"><term>Algorithms</term>
<term>Artificial Intelligence</term>
<term>Benchmarking</term>
<term>Computer Graphics</term>
<term>Models, Statistical</term>
<term>Numerical Analysis, Computer-Assisted</term>
<term>Reproducibility of Results</term>
<term>Sensitivity and Specificity</term>
<term>Signal Processing, Computer-Assisted</term>
<term>Subtraction Technique</term>
<term>User-Computer Interface</term>
</keywords>
<keywords scheme="Pascal" xml:lang="fr"><term>Intelligence artificielle</term>
<term>Analyse forme</term>
<term>Reconnaissance caractère</term>
<term>Reconnaissance optique caractère</term>
<term>Base de données</term>
<term>Traitement document</term>
<term>Evaluation performance</term>
<term>Réalité terrain</term>
<term>Analyse texte</term>
<term>Métrique</term>
<term>Segmentation</term>
<term>Optimisation</term>
<term>Diagramme Voronoï</term>
</keywords>
<keywords scheme="Wicri" type="topic" xml:lang="fr"><term>Intelligence artificielle</term>
<term>Base de données</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">-Informative benchmarks are crucial for optimizing the page segmentation step of an OCR system, frequently the performance limiting step for overall OCR system performance. We show that current evaluation scores are insufficient for diagnosing specific errors in page segmentation and fail to identify some classes of serious segmentation errors altogether. This paper introduces a vectorial score that is sensitive to, and identifies, the most important classes of segmentation errors (over, under, and mis-segmentation) and what page components (lines, blocks, etc.) are affected. Unlike previous schemes, our evaluation method has a canonical representation of ground-truth data and guarantees pixel-accurate evaluation results for arbitrary region shapes. We present the results of evaluating widely used segmentation algorithms (x-y cut, smearing, whitespace analysis, constrained text-line finding, docstrum, and Voronoi) on the UW-III database and demonstrate that the new evaluation scheme permits the identification of several specific flaws in individual segmentation methods.</div>
</front>
</TEI>
<affiliations><list><country><li>Allemagne</li>
</country>
<region><li>Rhénanie-Palatinat</li>
</region>
<settlement><li>Kaiserslautern</li>
</settlement>
<orgName><li>Université de technologie de Kaiserslautern</li>
</orgName>
</list>
<tree><country name="Allemagne"><region name="Rhénanie-Palatinat"><name sortKey="Shafait, Faisal" sort="Shafait, Faisal" uniqKey="Shafait F" first="Faisal" last="Shafait">Faisal Shafait</name>
</region>
<name sortKey="Breuel, Thomas M" sort="Breuel, Thomas M" uniqKey="Breuel T" first="Thomas M." last="Breuel">Thomas M. Breuel</name>
<name sortKey="Keysers, Daniel" sort="Keysers, Daniel" uniqKey="Keysers D" first="Daniel" last="Keysers">Daniel Keysers</name>
</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 000D07 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000D07 | 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:08-0254155 |texte= Performance Evaluation and Benchmarking of Six-Page Segmentation Algorithms }}
This area was generated with Dilib version V0.6.32. |