Textual summarisation of flowcharts in patent drawings for CLEF-IP 2012
Identifieur interne : 000120 ( Hal/Corpus ); précédent : 000119; suivant : 000121Textual summarisation of flowcharts in patent drawings for CLEF-IP 2012
Auteurs : Andrew Thean ; Jean-Marc Deltorn ; Patrice Lopez ; Laurent RomarySource :
Abstract
The CLEF-IP 2012 track included the Flowchart Recognition task, an image-based task where the goal was to process binary images of flowcharts taken from patent draw- ings to produce summaries containing information about their structure. The textual summaries include information about the flowchart title, the box-node shapes, the con- necting edge types, text describing flowchart content and the structural relationships between nodes and edges. An algorithm designed for this task and characterised by the following method steps is presented: * Text-graphic segmentation based on connected-component clustering; * Line segment bridging with an adaptive, oriented filter; * Box shape classification using a stretch-invariant transform to extract features based on shape-specific symmetry; * Text object recognition using a noisy channel model to enhance the results of a commercial OCR package. Performance evaluation results for the CLEF-IP 2012 Flowchart Recognition task are not yet available so the performance of the algorithm has been measured by com- paring algorithm output with object-level ground-truth values. An average F-score was calculated by combining node classification and edge detection (ignoring edge di- rectivity). Using this measure, a third of all drawings were recognized without error (average F-score=1.00) and 75% show an F-score of 0.78 or better. The most impor- tant failure modes of the algorithm have been identified as text-graphic segmentation, line-segment bridging and edge directivity classification. The text object recognition module of the algorithm has been independently eval- uated. Two different state-of-the-art OCR software packages were compared and a post-correction method was applied to their output. Post-correction yields an im- provement of 9% in OCR accuracy and a 26% reduction in the word error rate.
Url:
Links to Exploration step
Hal:hal-00728779Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Textual summarisation of flowcharts in patent drawings for CLEF-IP 2012</title><author><name sortKey="Thean, Andrew" sort="Thean, Andrew" uniqKey="Thean A" first="Andrew" last="Thean">Andrew Thean</name><affiliation><hal:affiliation type="laboratory" xml:id="struct-95237" status="VALID"><orgName>Institut für Deutsche Sprache und Linguistik</orgName><orgName type="acronym">IDSL</orgName><desc><address><addrLine>Dorotheenstraße 24, 10099 Berlin</addrLine><country key="DE"></country></address><ref type="url">http://www.linguistik.hu-berlin.de/</ref></desc><listRelation><relation active="#struct-139189" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-139189" type="direct"><org type="institution" xml:id="struct-139189" status="VALID"><orgName>Humboldt Universität zu Berlin [Berlin]</orgName><desc><address><addrLine>Unter den Linden 610099 Berlin</addrLine><country key="DE"></country></address><ref type="url">https://www.hu-berlin.de/en/</ref></desc></org></tutelle></tutelles></hal:affiliation></affiliation></author><author><name sortKey="Deltorn, Jean Marc" sort="Deltorn, Jean Marc" uniqKey="Deltorn J" first="Jean-Marc" last="Deltorn">Jean-Marc Deltorn</name><affiliation><hal:affiliation type="laboratory" xml:id="struct-95237" status="VALID"><orgName>Institut für Deutsche Sprache und Linguistik</orgName><orgName type="acronym">IDSL</orgName><desc><address><addrLine>Dorotheenstraße 24, 10099 Berlin</addrLine><country key="DE"></country></address><ref type="url">http://www.linguistik.hu-berlin.de/</ref></desc><listRelation><relation active="#struct-139189" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-139189" type="direct"><org type="institution" xml:id="struct-139189" status="VALID"><orgName>Humboldt Universität zu Berlin [Berlin]</orgName><desc><address><addrLine>Unter den Linden 610099 Berlin</addrLine><country key="DE"></country></address><ref type="url">https://www.hu-berlin.de/en/</ref></desc></org></tutelle></tutelles></hal:affiliation></affiliation></author><author><name sortKey="Lopez, Patrice" sort="Lopez, Patrice" uniqKey="Lopez P" first="Patrice" last="Lopez">Patrice Lopez</name><affiliation><hal:affiliation type="laboratory" xml:id="struct-95237" status="VALID"><orgName>Institut für Deutsche Sprache und Linguistik</orgName><orgName type="acronym">IDSL</orgName><desc><address><addrLine>Dorotheenstraße 24, 10099 Berlin</addrLine><country key="DE"></country></address><ref type="url">http://www.linguistik.hu-berlin.de/</ref></desc><listRelation><relation active="#struct-139189" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-139189" type="direct"><org type="institution" xml:id="struct-139189" status="VALID"><orgName>Humboldt Universität zu Berlin [Berlin]</orgName><desc><address><addrLine>Unter den Linden 610099 Berlin</addrLine><country key="DE"></country></address><ref type="url">https://www.hu-berlin.de/en/</ref></desc></org></tutelle></tutelles></hal:affiliation></affiliation></author><author><name sortKey="Romary, Laurent" sort="Romary, Laurent" uniqKey="Romary L" first="Laurent" last="Romary">Laurent Romary</name><affiliation><hal:affiliation type="laboratory" xml:id="struct-95237" status="VALID"><orgName>Institut für Deutsche Sprache und Linguistik</orgName><orgName type="acronym">IDSL</orgName><desc><address><addrLine>Dorotheenstraße 24, 10099 Berlin</addrLine><country key="DE"></country></address><ref type="url">http://www.linguistik.hu-berlin.de/</ref></desc><listRelation><relation active="#struct-139189" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-139189" type="direct"><org type="institution" xml:id="struct-139189" status="VALID"><orgName>Humboldt Universität zu Berlin [Berlin]</orgName><desc><address><addrLine>Unter den Linden 610099 Berlin</addrLine><country key="DE"></country></address><ref type="url">https://www.hu-berlin.de/en/</ref></desc></org></tutelle></tutelles></hal:affiliation></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">HAL</idno><idno type="RBID">Hal:hal-00728779</idno><idno type="halId">hal-00728779</idno><idno type="halUri">https://hal.inria.fr/hal-00728779</idno><idno type="url">https://hal.inria.fr/hal-00728779</idno><date when="2012-09-17">2012-09-17</date><idno type="wicri:Area/Hal/Corpus">000120</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">Textual summarisation of flowcharts in patent drawings for CLEF-IP 2012</title><author><name sortKey="Thean, Andrew" sort="Thean, Andrew" uniqKey="Thean A" first="Andrew" last="Thean">Andrew Thean</name><affiliation><hal:affiliation type="laboratory" xml:id="struct-95237" status="VALID"><orgName>Institut für Deutsche Sprache und Linguistik</orgName><orgName type="acronym">IDSL</orgName><desc><address><addrLine>Dorotheenstraße 24, 10099 Berlin</addrLine><country key="DE"></country></address><ref type="url">http://www.linguistik.hu-berlin.de/</ref></desc><listRelation><relation active="#struct-139189" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-139189" type="direct"><org type="institution" xml:id="struct-139189" status="VALID"><orgName>Humboldt Universität zu Berlin [Berlin]</orgName><desc><address><addrLine>Unter den Linden 610099 Berlin</addrLine><country key="DE"></country></address><ref type="url">https://www.hu-berlin.de/en/</ref></desc></org></tutelle></tutelles></hal:affiliation></affiliation></author><author><name sortKey="Deltorn, Jean Marc" sort="Deltorn, Jean Marc" uniqKey="Deltorn J" first="Jean-Marc" last="Deltorn">Jean-Marc Deltorn</name><affiliation><hal:affiliation type="laboratory" xml:id="struct-95237" status="VALID"><orgName>Institut für Deutsche Sprache und Linguistik</orgName><orgName type="acronym">IDSL</orgName><desc><address><addrLine>Dorotheenstraße 24, 10099 Berlin</addrLine><country key="DE"></country></address><ref type="url">http://www.linguistik.hu-berlin.de/</ref></desc><listRelation><relation active="#struct-139189" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-139189" type="direct"><org type="institution" xml:id="struct-139189" status="VALID"><orgName>Humboldt Universität zu Berlin [Berlin]</orgName><desc><address><addrLine>Unter den Linden 610099 Berlin</addrLine><country key="DE"></country></address><ref type="url">https://www.hu-berlin.de/en/</ref></desc></org></tutelle></tutelles></hal:affiliation></affiliation></author><author><name sortKey="Lopez, Patrice" sort="Lopez, Patrice" uniqKey="Lopez P" first="Patrice" last="Lopez">Patrice Lopez</name><affiliation><hal:affiliation type="laboratory" xml:id="struct-95237" status="VALID"><orgName>Institut für Deutsche Sprache und Linguistik</orgName><orgName type="acronym">IDSL</orgName><desc><address><addrLine>Dorotheenstraße 24, 10099 Berlin</addrLine><country key="DE"></country></address><ref type="url">http://www.linguistik.hu-berlin.de/</ref></desc><listRelation><relation active="#struct-139189" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-139189" type="direct"><org type="institution" xml:id="struct-139189" status="VALID"><orgName>Humboldt Universität zu Berlin [Berlin]</orgName><desc><address><addrLine>Unter den Linden 610099 Berlin</addrLine><country key="DE"></country></address><ref type="url">https://www.hu-berlin.de/en/</ref></desc></org></tutelle></tutelles></hal:affiliation></affiliation></author><author><name sortKey="Romary, Laurent" sort="Romary, Laurent" uniqKey="Romary L" first="Laurent" last="Romary">Laurent Romary</name><affiliation><hal:affiliation type="laboratory" xml:id="struct-95237" status="VALID"><orgName>Institut für Deutsche Sprache und Linguistik</orgName><orgName type="acronym">IDSL</orgName><desc><address><addrLine>Dorotheenstraße 24, 10099 Berlin</addrLine><country key="DE"></country></address><ref type="url">http://www.linguistik.hu-berlin.de/</ref></desc><listRelation><relation active="#struct-139189" type="direct"></relation></listRelation><tutelles><tutelle active="#struct-139189" type="direct"><org type="institution" xml:id="struct-139189" status="VALID"><orgName>Humboldt Universität zu Berlin [Berlin]</orgName><desc><address><addrLine>Unter den Linden 610099 Berlin</addrLine><country key="DE"></country></address><ref type="url">https://www.hu-berlin.de/en/</ref></desc></org></tutelle></tutelles></hal:affiliation></affiliation></author></analytic></biblStruct></sourceDesc></fileDesc><profileDesc><textClass></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">The CLEF-IP 2012 track included the Flowchart Recognition task, an image-based task where the goal was to process binary images of flowcharts taken from patent draw- ings to produce summaries containing information about their structure. The textual summaries include information about the flowchart title, the box-node shapes, the con- necting edge types, text describing flowchart content and the structural relationships between nodes and edges. An algorithm designed for this task and characterised by the following method steps is presented: * Text-graphic segmentation based on connected-component clustering; * Line segment bridging with an adaptive, oriented filter; * Box shape classification using a stretch-invariant transform to extract features based on shape-specific symmetry; * Text object recognition using a noisy channel model to enhance the results of a commercial OCR package. Performance evaluation results for the CLEF-IP 2012 Flowchart Recognition task are not yet available so the performance of the algorithm has been measured by com- paring algorithm output with object-level ground-truth values. An average F-score was calculated by combining node classification and edge detection (ignoring edge di- rectivity). Using this measure, a third of all drawings were recognized without error (average F-score=1.00) and 75% show an F-score of 0.78 or better. The most impor- tant failure modes of the algorithm have been identified as text-graphic segmentation, line-segment bridging and edge directivity classification. The text object recognition module of the algorithm has been independently eval- uated. Two different state-of-the-art OCR software packages were compared and a post-correction method was applied to their output. Post-correction yields an im- provement of 9% in OCR accuracy and a 26% reduction in the word error rate.</div></front></TEI><hal api="V3"><titleStmt><title xml:lang="en">Textual summarisation of flowcharts in patent drawings for CLEF-IP 2012</title><author role="aut"><persName><forename type="first">Andrew</forename><surname>Thean</surname></persName><email></email><idno type="halauthor">758449</idno><affiliation ref="#struct-95237"></affiliation><affiliation ref="#struct-118511"></affiliation></author><author role="aut"><persName><forename type="first">Jean-Marc</forename><surname>Deltorn</surname></persName><email></email><idno type="halauthor">758450</idno><affiliation ref="#struct-95237"></affiliation><affiliation ref="#struct-118511"></affiliation></author><author role="aut"><persName><forename type="first">Patrice</forename><surname>Lopez</surname></persName><email></email><idno type="idhal">patricelopez</idno><idno type="halauthor">144548</idno><affiliation ref="#struct-95237"></affiliation><affiliation ref="#struct-118511"></affiliation></author><author role="aut"><persName><forename type="first">Laurent</forename><surname>Romary</surname></persName><email>laurent.romary@inria.fr</email><idno type="idhal">laurentromary</idno><idno type="halauthor">49567</idno><idno type="arXiv">http://arxiv.org/a/Romary_L</idno><idno type="IdRef">http://www.idref.fr/060702494</idno><idno type="ORCID">http://orcid.org/0000-0002-0756-0508</idno><idno type="VIAF">http://viaf.org/viaf/VIAF282014122</idno><idno type="ISNI">http://isni.org/isni/0000 0003 8879 5444</idno><affiliation ref="#struct-95237"></affiliation><affiliation ref="#struct-118511"></affiliation></author><editor role="depositor"><persName><forename>Laurent</forename><surname>Romary</surname></persName><email>laurent.romary@inria.fr</email></editor></titleStmt><editionStmt><edition n="v1" type="current"><date type="whenSubmitted">2012-09-06 15:55:11</date><date type="whenModified">2012-09-07 09:39:57</date><date type="whenReleased">2012-09-07 09:39:57</date><date type="whenProduced">2012-09-17</date><date type="whenEndEmbargoed">2012-09-06</date><ref type="file" target="https://hal.inria.fr/hal-00728779/document"><date notBefore="2012-09-06"></date></ref><ref type="file" subtype="author" n="1" target="https://hal.inria.fr/hal-00728779/file/clef-ip-2012-flow.pdf"><date notBefore="2012-09-06"></date></ref></edition><respStmt><resp>contributor</resp><name key="105529"><persName><forename>Laurent</forename><surname>Romary</surname></persName><email>laurent.romary@inria.fr</email></name></respStmt></editionStmt><publicationStmt><distributor>CCSD</distributor><idno type="halId">hal-00728779</idno><idno type="halUri">https://hal.inria.fr/hal-00728779</idno><idno type="halBibtex">thean:hal-00728779</idno><idno type="halRefHtml">CLEF 2012, Sep 2012, Roma, Italy. 2012</idno><idno type="halRef">CLEF 2012, Sep 2012, Roma, Italy. 2012</idno></publicationStmt><seriesStmt><idno type="stamp" n="INRIA">INRIA - Institut National de Recherche en Informatique et en Automatique</idno><idno type="stamp" n="INRIA-SACLAY">INRIA Saclay - Ile de France</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">Textual summarisation of flowcharts in patent drawings for CLEF-IP 2012</title><author role="aut"><persName><forename type="first">Andrew</forename><surname>Thean</surname></persName><idno type="halAuthorId">758449</idno><affiliation ref="#struct-95237"></affiliation><affiliation ref="#struct-118511"></affiliation></author><author role="aut"><persName><forename type="first">Jean-Marc</forename><surname>Deltorn</surname></persName><idno type="halAuthorId">758450</idno><affiliation ref="#struct-95237"></affiliation><affiliation ref="#struct-118511"></affiliation></author><author role="aut"><persName><forename type="first">Patrice</forename><surname>Lopez</surname></persName><idno type="idHal">patricelopez</idno><idno type="halAuthorId">144548</idno><affiliation ref="#struct-95237"></affiliation><affiliation ref="#struct-118511"></affiliation></author><author role="aut"><persName><forename type="first">Laurent</forename><surname>Romary</surname></persName><email>laurent.romary@inria.fr</email><idno type="idHal">laurentromary</idno><idno type="halAuthorId">49567</idno><idno type="arXiv">http://arxiv.org/a/Romary_L</idno><idno type="IdRef">http://www.idref.fr/060702494</idno><idno type="ORCID">http://orcid.org/0000-0002-0756-0508</idno><idno type="VIAF">http://viaf.org/viaf/VIAF282014122</idno><idno type="ISNI">http://isni.org/isni/0000 0003 8879 5444</idno><affiliation ref="#struct-95237"></affiliation><affiliation ref="#struct-118511"></affiliation></author></analytic><monogr><meeting><title>CLEF 2012</title><date type="start">2012-09-17</date><date type="end">2012-09-20</date><settlement>Roma</settlement><country key="IT">Italy</country></meeting><imprint><date type="datePub">2012</date></imprint></monogr></biblStruct></sourceDesc><profileDesc><langUsage><language ident="en">English</language></langUsage><textClass><classCode scheme="classification">H.3 [Information Storage and Retrieval]: H.3.1 Content Analysis and Indexing; H.3.3 Infor- mation Search and Retrieval; H.3.4 Systems and Software; H.3.7 Digital Libraries</classCode><classCode scheme="halDomain" n="info.info-cl">Computer Science [cs]/Computation and Language [cs.CL]</classCode><classCode scheme="halTypology" n="COMM">Conference papers</classCode></textClass><abstract xml:lang="en">The CLEF-IP 2012 track included the Flowchart Recognition task, an image-based task where the goal was to process binary images of flowcharts taken from patent draw- ings to produce summaries containing information about their structure. The textual summaries include information about the flowchart title, the box-node shapes, the con- necting edge types, text describing flowchart content and the structural relationships between nodes and edges. An algorithm designed for this task and characterised by the following method steps is presented: * Text-graphic segmentation based on connected-component clustering; * Line segment bridging with an adaptive, oriented filter; * Box shape classification using a stretch-invariant transform to extract features based on shape-specific symmetry; * Text object recognition using a noisy channel model to enhance the results of a commercial OCR package. Performance evaluation results for the CLEF-IP 2012 Flowchart Recognition task are not yet available so the performance of the algorithm has been measured by com- paring algorithm output with object-level ground-truth values. An average F-score was calculated by combining node classification and edge detection (ignoring edge di- rectivity). Using this measure, a third of all drawings were recognized without error (average F-score=1.00) and 75% show an F-score of 0.78 or better. The most impor- tant failure modes of the algorithm have been identified as text-graphic segmentation, line-segment bridging and edge directivity classification. The text object recognition module of the algorithm has been independently eval- uated. Two different state-of-the-art OCR software packages were compared and a post-correction method was applied to their output. Post-correction yields an im- provement of 9% in OCR accuracy and a 26% reduction in the word error rate.</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 000120 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Hal/Corpus/biblio.hfd -nk 000120 | 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-00728779
|texte= Textual summarisation of flowcharts in patent drawings for CLEF-IP 2012
}}
| This area was generated with Dilib version V0.6.32. |  |