Word spotting in historical printed documents using shape and sequence comparisons
Identifieur interne : 000056 ( PascalFrancis/Checkpoint ); précédent : 000055; suivant : 000057Word spotting in historical printed documents using shape and sequence comparisons
Auteurs : Khurram Khurshid [France, Pakistan] ; Claudie Faure [France] ; Nicole Vincent [France]Source :
- Pattern recognition [ 0031-3203 ] ; 2012.
Descripteurs français
- Pascal (Inist)
- Wicri :
- topic : Prototype.
English descriptors
- KwdEn :
Abstract
Information spotting in scanned historical document images is a very challenging task. The joint use of the mechanical press and of human controlled inking introduced great variability in ink level within a book or even within a page. Consequently characters are often broken or merged together and thus become difficult to segment and recognize. The limitations of commercial OCR engines for information retrieval in historical document images have inspired alternative means of identification of given words in such documents. We present a word spotting method for scanned documents in order to find the word images that are similar to a query word, without assuming a correct segmentation of the words into characters. The connected components are first processed to transform a word pattern into a sequence of sub-patterns. Each sub-pattern is represented by a sequence of feature vectors. A modified Edit distance is proposed to perform a segmentation-driven string matching and to compute the Segmentation Driven Edit (SDE) distance between the words to be compared. The set of SDE operations is defined to obtain the word segmentations that are the most appropriate to evaluate their similarity. These operations are efficient to cope with broken and touching characters in words. The distortion of character shapes is handled by coupling the string matching process with local shape comparisons that are achieved by Dynamic Time Warping (DTW). The costs of the SDE operations are provided by the DTW distances. A sub-optimal version of the SDE string matching is also proposed to reduce the computation time, nevertheless it did not lead to a great decrease in performance. It is possible to enter a query by example or a textual query entered with the keyboard. Textual queries can be used to directly spot the word without the need to synthesize its image, as far as character prototype images are available. Results are presented for different documents and compared with other methods, showing the efficiency of our method.
Affiliations:
Links toward previous steps (curation, corpus...)
Links to Exploration step
Pascal:12-0159268Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">Word spotting in historical printed documents using shape and sequence comparisons</title>
<author><name sortKey="Khurshid, Khurram" sort="Khurshid, Khurram" uniqKey="Khurshid K" first="Khurram" last="Khurshid">Khurram Khurshid</name>
<affiliation wicri:level="3"><inist:fA14 i1="01"><s1>LIPADE-SIP, University Paris Descartes, 45, Rue des Saint Pères</s1>
<s2>75006 Paris</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>France</country>
<placeName><region type="region" nuts="2">Île-de-France</region>
<settlement type="city">Paris</settlement>
</placeName>
</affiliation>
<affiliation wicri:level="1"><inist:fA14 i1="03"><s1>Institute of Space Technology</s1>
<s2>Islamabad</s2>
<s3>PAK</s3>
<sZ>1 aut.</sZ>
</inist:fA14>
<country>Pakistan</country>
<wicri:noRegion>Institute of Space Technology</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Faure, Claudie" sort="Faure, Claudie" uniqKey="Faure C" first="Claudie" last="Faure">Claudie Faure</name>
<affiliation wicri:level="3"><inist:fA14 i1="02"><s1>LTCI CNRS-Telecom-ParisTech, 46, Rue Barrault</s1>
<s2>75634 Paris</s2>
<s3>FRA</s3>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>France</country>
<placeName><region type="region" nuts="2">Île-de-France</region>
<settlement type="city">Paris</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Vincent, Nicole" sort="Vincent, Nicole" uniqKey="Vincent N" first="Nicole" last="Vincent">Nicole Vincent</name>
<affiliation wicri:level="3"><inist:fA14 i1="01"><s1>LIPADE-SIP, University Paris Descartes, 45, Rue des Saint Pères</s1>
<s2>75006 Paris</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>France</country>
<placeName><region type="region" nuts="2">Île-de-France</region>
<settlement type="city">Paris</settlement>
</placeName>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">INIST</idno>
<idno type="inist">12-0159268</idno>
<date when="2012">2012</date>
<idno type="stanalyst">PASCAL 12-0159268 INIST</idno>
<idno type="RBID">Pascal:12-0159268</idno>
<idno type="wicri:Area/PascalFrancis/Corpus">000097</idno>
<idno type="wicri:Area/PascalFrancis/Curation">000675</idno>
<idno type="wicri:Area/PascalFrancis/Checkpoint">000056</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">Word spotting in historical printed documents using shape and sequence comparisons</title>
<author><name sortKey="Khurshid, Khurram" sort="Khurshid, Khurram" uniqKey="Khurshid K" first="Khurram" last="Khurshid">Khurram Khurshid</name>
<affiliation wicri:level="3"><inist:fA14 i1="01"><s1>LIPADE-SIP, University Paris Descartes, 45, Rue des Saint Pères</s1>
<s2>75006 Paris</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>France</country>
<placeName><region type="region" nuts="2">Île-de-France</region>
<settlement type="city">Paris</settlement>
</placeName>
</affiliation>
<affiliation wicri:level="1"><inist:fA14 i1="03"><s1>Institute of Space Technology</s1>
<s2>Islamabad</s2>
<s3>PAK</s3>
<sZ>1 aut.</sZ>
</inist:fA14>
<country>Pakistan</country>
<wicri:noRegion>Institute of Space Technology</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Faure, Claudie" sort="Faure, Claudie" uniqKey="Faure C" first="Claudie" last="Faure">Claudie Faure</name>
<affiliation wicri:level="3"><inist:fA14 i1="02"><s1>LTCI CNRS-Telecom-ParisTech, 46, Rue Barrault</s1>
<s2>75634 Paris</s2>
<s3>FRA</s3>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>France</country>
<placeName><region type="region" nuts="2">Île-de-France</region>
<settlement type="city">Paris</settlement>
</placeName>
</affiliation>
</author>
<author><name sortKey="Vincent, Nicole" sort="Vincent, Nicole" uniqKey="Vincent N" first="Nicole" last="Vincent">Nicole Vincent</name>
<affiliation wicri:level="3"><inist:fA14 i1="01"><s1>LIPADE-SIP, University Paris Descartes, 45, Rue des Saint Pères</s1>
<s2>75006 Paris</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>France</country>
<placeName><region type="region" nuts="2">Île-de-France</region>
<settlement type="city">Paris</settlement>
</placeName>
</affiliation>
</author>
</analytic>
<series><title level="j" type="main">Pattern recognition</title>
<title level="j" type="abbreviated">Pattern recogn.</title>
<idno type="ISSN">0031-3203</idno>
<imprint><date when="2012">2012</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><title level="j" type="main">Pattern recognition</title>
<title level="j" type="abbreviated">Pattern recogn.</title>
<idno type="ISSN">0031-3203</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Computation time</term>
<term>Document image processing</term>
<term>Electronic library</term>
<term>Information retrieval</term>
<term>Optical character recognition</term>
<term>Pattern recognition</term>
<term>Performance evaluation</term>
<term>Printed document</term>
<term>Prototype</term>
<term>Query</term>
<term>Segmentation</term>
<term>Similarity</term>
<term>Time warp simulation</term>
</keywords>
<keywords scheme="Pascal" xml:lang="fr"><term>Document imprimé</term>
<term>Traitement image document</term>
<term>Reconnaissance optique caractère</term>
<term>Recherche information</term>
<term>Requête</term>
<term>Segmentation</term>
<term>Similitude</term>
<term>Simulation Time Warp</term>
<term>Temps calcul</term>
<term>Evaluation performance</term>
<term>Prototype</term>
<term>Bibliothèque électronique</term>
<term>Reconnaissance forme</term>
</keywords>
<keywords scheme="Wicri" type="topic" xml:lang="fr"><term>Prototype</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">Information spotting in scanned historical document images is a very challenging task. The joint use of the mechanical press and of human controlled inking introduced great variability in ink level within a book or even within a page. Consequently characters are often broken or merged together and thus become difficult to segment and recognize. The limitations of commercial OCR engines for information retrieval in historical document images have inspired alternative means of identification of given words in such documents. We present a word spotting method for scanned documents in order to find the word images that are similar to a query word, without assuming a correct segmentation of the words into characters. The connected components are first processed to transform a word pattern into a sequence of sub-patterns. Each sub-pattern is represented by a sequence of feature vectors. A modified Edit distance is proposed to perform a segmentation-driven string matching and to compute the Segmentation Driven Edit (SDE) distance between the words to be compared. The set of SDE operations is defined to obtain the word segmentations that are the most appropriate to evaluate their similarity. These operations are efficient to cope with broken and touching characters in words. The distortion of character shapes is handled by coupling the string matching process with local shape comparisons that are achieved by Dynamic Time Warping (DTW). The costs of the SDE operations are provided by the DTW distances. A sub-optimal version of the SDE string matching is also proposed to reduce the computation time, nevertheless it did not lead to a great decrease in performance. It is possible to enter a query by example or a textual query entered with the keyboard. Textual queries can be used to directly spot the word without the need to synthesize its image, as far as character prototype images are available. Results are presented for different documents and compared with other methods, showing the efficiency of our method.</div>
</front>
</TEI>
<inist><standard h6="B"><pA><fA01 i1="01" i2="1"><s0>0031-3203</s0>
</fA01>
<fA02 i1="01"><s0>PTNRA8</s0>
</fA02>
<fA03 i2="1"><s0>Pattern recogn.</s0>
</fA03>
<fA05><s2>45</s2>
</fA05>
<fA06><s2>7</s2>
</fA06>
<fA08 i1="01" i2="1" l="ENG"><s1>Word spotting in historical printed documents using shape and sequence comparisons</s1>
</fA08>
<fA11 i1="01" i2="1"><s1>KHURSHID (Khurram)</s1>
</fA11>
<fA11 i1="02" i2="1"><s1>FAURE (Claudie)</s1>
</fA11>
<fA11 i1="03" i2="1"><s1>VINCENT (Nicole)</s1>
</fA11>
<fA14 i1="01"><s1>LIPADE-SIP, University Paris Descartes, 45, Rue des Saint Pères</s1>
<s2>75006 Paris</s2>
<s3>FRA</s3>
<sZ>1 aut.</sZ>
<sZ>3 aut.</sZ>
</fA14>
<fA14 i1="02"><s1>LTCI CNRS-Telecom-ParisTech, 46, Rue Barrault</s1>
<s2>75634 Paris</s2>
<s3>FRA</s3>
<sZ>2 aut.</sZ>
</fA14>
<fA14 i1="03"><s1>Institute of Space Technology</s1>
<s2>Islamabad</s2>
<s3>PAK</s3>
<sZ>1 aut.</sZ>
</fA14>
<fA20><s1>2598-2609</s1>
</fA20>
<fA21><s1>2012</s1>
</fA21>
<fA23 i1="01"><s0>ENG</s0>
</fA23>
<fA43 i1="01"><s1>INIST</s1>
<s2>15220</s2>
<s5>354000509613340130</s5>
</fA43>
<fA44><s0>0000</s0>
<s1>© 2012 INIST-CNRS. All rights reserved.</s1>
</fA44>
<fA45><s0>41 ref.</s0>
</fA45>
<fA47 i1="01" i2="1"><s0>12-0159268</s0>
</fA47>
<fA60><s1>P</s1>
</fA60>
<fA61><s0>A</s0>
</fA61>
<fA64 i1="01" i2="1"><s0>Pattern recognition</s0>
</fA64>
<fA66 i1="01"><s0>GBR</s0>
</fA66>
<fC01 i1="01" l="ENG"><s0>Information spotting in scanned historical document images is a very challenging task. The joint use of the mechanical press and of human controlled inking introduced great variability in ink level within a book or even within a page. Consequently characters are often broken or merged together and thus become difficult to segment and recognize. The limitations of commercial OCR engines for information retrieval in historical document images have inspired alternative means of identification of given words in such documents. We present a word spotting method for scanned documents in order to find the word images that are similar to a query word, without assuming a correct segmentation of the words into characters. The connected components are first processed to transform a word pattern into a sequence of sub-patterns. Each sub-pattern is represented by a sequence of feature vectors. A modified Edit distance is proposed to perform a segmentation-driven string matching and to compute the Segmentation Driven Edit (SDE) distance between the words to be compared. The set of SDE operations is defined to obtain the word segmentations that are the most appropriate to evaluate their similarity. These operations are efficient to cope with broken and touching characters in words. The distortion of character shapes is handled by coupling the string matching process with local shape comparisons that are achieved by Dynamic Time Warping (DTW). The costs of the SDE operations are provided by the DTW distances. A sub-optimal version of the SDE string matching is also proposed to reduce the computation time, nevertheless it did not lead to a great decrease in performance. It is possible to enter a query by example or a textual query entered with the keyboard. Textual queries can be used to directly spot the word without the need to synthesize its image, as far as character prototype images are available. Results are presented for different documents and compared with other methods, showing the efficiency of our method.</s0>
</fC01>
<fC02 i1="01" i2="X"><s0>001D04A05C</s0>
</fC02>
<fC02 i1="02" i2="X"><s0>001D04A05A</s0>
</fC02>
<fC02 i1="03" i2="X"><s0>001D04A03</s0>
</fC02>
<fC03 i1="01" i2="X" l="FRE"><s0>Document imprimé</s0>
<s5>01</s5>
</fC03>
<fC03 i1="01" i2="X" l="ENG"><s0>Printed document</s0>
<s5>01</s5>
</fC03>
<fC03 i1="01" i2="X" l="SPA"><s0>Documento impreso</s0>
<s5>01</s5>
</fC03>
<fC03 i1="02" i2="3" l="FRE"><s0>Traitement image document</s0>
<s5>02</s5>
</fC03>
<fC03 i1="02" i2="3" l="ENG"><s0>Document image processing</s0>
<s5>02</s5>
</fC03>
<fC03 i1="03" i2="X" l="FRE"><s0>Reconnaissance optique caractère</s0>
<s5>03</s5>
</fC03>
<fC03 i1="03" i2="X" l="ENG"><s0>Optical character recognition</s0>
<s5>03</s5>
</fC03>
<fC03 i1="03" i2="X" l="SPA"><s0>Reconocimento óptico de caracteres</s0>
<s5>03</s5>
</fC03>
<fC03 i1="04" i2="X" l="FRE"><s0>Recherche information</s0>
<s5>04</s5>
</fC03>
<fC03 i1="04" i2="X" l="ENG"><s0>Information retrieval</s0>
<s5>04</s5>
</fC03>
<fC03 i1="04" i2="X" l="SPA"><s0>Búsqueda información</s0>
<s5>04</s5>
</fC03>
<fC03 i1="05" i2="X" l="FRE"><s0>Requête</s0>
<s5>05</s5>
</fC03>
<fC03 i1="05" i2="X" l="ENG"><s0>Query</s0>
<s5>05</s5>
</fC03>
<fC03 i1="05" i2="X" l="SPA"><s0>Pregunta documental</s0>
<s5>05</s5>
</fC03>
<fC03 i1="06" i2="X" l="FRE"><s0>Segmentation</s0>
<s5>06</s5>
</fC03>
<fC03 i1="06" i2="X" l="ENG"><s0>Segmentation</s0>
<s5>06</s5>
</fC03>
<fC03 i1="06" i2="X" l="SPA"><s0>Segmentación</s0>
<s5>06</s5>
</fC03>
<fC03 i1="07" i2="X" l="FRE"><s0>Similitude</s0>
<s5>07</s5>
</fC03>
<fC03 i1="07" i2="X" l="ENG"><s0>Similarity</s0>
<s5>07</s5>
</fC03>
<fC03 i1="07" i2="X" l="SPA"><s0>Similitud</s0>
<s5>07</s5>
</fC03>
<fC03 i1="08" i2="3" l="FRE"><s0>Simulation Time Warp</s0>
<s5>08</s5>
</fC03>
<fC03 i1="08" i2="3" l="ENG"><s0>Time warp simulation</s0>
<s5>08</s5>
</fC03>
<fC03 i1="09" i2="X" l="FRE"><s0>Temps calcul</s0>
<s5>09</s5>
</fC03>
<fC03 i1="09" i2="X" l="ENG"><s0>Computation time</s0>
<s5>09</s5>
</fC03>
<fC03 i1="09" i2="X" l="SPA"><s0>Tiempo computación</s0>
<s5>09</s5>
</fC03>
<fC03 i1="10" i2="X" l="FRE"><s0>Evaluation performance</s0>
<s5>10</s5>
</fC03>
<fC03 i1="10" i2="X" l="ENG"><s0>Performance evaluation</s0>
<s5>10</s5>
</fC03>
<fC03 i1="10" i2="X" l="SPA"><s0>Evaluación prestación</s0>
<s5>10</s5>
</fC03>
<fC03 i1="11" i2="X" l="FRE"><s0>Prototype</s0>
<s5>11</s5>
</fC03>
<fC03 i1="11" i2="X" l="ENG"><s0>Prototype</s0>
<s5>11</s5>
</fC03>
<fC03 i1="11" i2="X" l="SPA"><s0>Prototipo</s0>
<s5>11</s5>
</fC03>
<fC03 i1="12" i2="X" l="FRE"><s0>Bibliothèque électronique</s0>
<s5>12</s5>
</fC03>
<fC03 i1="12" i2="X" l="ENG"><s0>Electronic library</s0>
<s5>12</s5>
</fC03>
<fC03 i1="12" i2="X" l="SPA"><s0>Biblioteca electronica</s0>
<s5>12</s5>
</fC03>
<fC03 i1="13" i2="X" l="FRE"><s0>Reconnaissance forme</s0>
<s5>46</s5>
</fC03>
<fC03 i1="13" i2="X" l="ENG"><s0>Pattern recognition</s0>
<s5>46</s5>
</fC03>
<fC03 i1="13" i2="X" l="SPA"><s0>Reconocimiento patrón</s0>
<s5>46</s5>
</fC03>
<fC07 i1="01" i2="X" l="FRE"><s0>Traitement image</s0>
<s5>13</s5>
</fC07>
<fC07 i1="01" i2="X" l="ENG"><s0>Image processing</s0>
<s5>13</s5>
</fC07>
<fC07 i1="01" i2="X" l="SPA"><s0>Procesamiento imagen</s0>
<s5>13</s5>
</fC07>
<fN21><s1>122</s1>
</fN21>
<fN44 i1="01"><s1>OTO</s1>
</fN44>
<fN82><s1>OTO</s1>
</fN82>
</pA>
</standard>
</inist>
<affiliations><list><country><li>France</li>
<li>Pakistan</li>
</country>
<region><li>Île-de-France</li>
</region>
<settlement><li>Paris</li>
</settlement>
</list>
<tree><country name="France"><region name="Île-de-France"><name sortKey="Khurshid, Khurram" sort="Khurshid, Khurram" uniqKey="Khurshid K" first="Khurram" last="Khurshid">Khurram Khurshid</name>
</region>
<name sortKey="Faure, Claudie" sort="Faure, Claudie" uniqKey="Faure C" first="Claudie" last="Faure">Claudie Faure</name>
<name sortKey="Vincent, Nicole" sort="Vincent, Nicole" uniqKey="Vincent N" first="Nicole" last="Vincent">Nicole Vincent</name>
</country>
<country name="Pakistan"><noRegion><name sortKey="Khurshid, Khurram" sort="Khurshid, Khurram" uniqKey="Khurshid K" first="Khurram" last="Khurshid">Khurram Khurshid</name>
</noRegion>
</country>
</tree>
</affiliations>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/PascalFrancis/Checkpoint
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000056 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/PascalFrancis/Checkpoint/biblio.hfd -nk 000056 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Ticri/CIDE |area= OcrV1 |flux= PascalFrancis |étape= Checkpoint |type= RBID |clé= Pascal:12-0159268 |texte= Word spotting in historical printed documents using shape and sequence comparisons }}
This area was generated with Dilib version V0.6.32. |