Text String Detection from Natural Scenes by Structure-based Partition and Grouping
Identifieur interne : 000145 ( Pmc/Curation ); précédent : 000144; suivant : 000146Text String Detection from Natural Scenes by Structure-based Partition and Grouping
Auteurs : Chucai Yi ; Yingli TianSource :
- Ieee Transactions on Image Processing [ 1057-7149 ] ; 2011.
Abstract
Text information in natural scene images serves as important clues for many image-based applications such as scene understanding, content-based image retrieval, assistive navigation, and automatic geocoding. However, locating text from complex background with multiple colors is a challenging task. In this paper, we explore a new framework to detect text strings with arbitrary orientations in complex natural scene images. Our proposed framework of text string detection consists of two steps: 1) Image partition to find text character candidates based on local gradient features and color uniformity of character components. 2) Character candidate grouping to detect text strings based on joint structural features of text characters in each text string such as character size differences, distances between neighboring characters, and character alignment. By assuming that a text string has at least three characters, we propose two algorithms of text string detection: 1) adjacent character grouping method, and 2) text line grouping method. The adjacent character grouping method calculates the sibling groups of each character candidate as string segments and then merges the intersecting sibling groups into text string. The text line grouping method performs Hough transform to fit text line among the centroids of text candidates. Each fitted text line describes the orientation of a potential text string. The detected text string is presented by a rectangle region covering all characters whose centroids are cascaded in its text line. To improve efficiency and accuracy, our algorithms are carried out in multi-scales. The proposed methods outperform the state-of-the-art results on the public Robust Reading Dataset which contains text only in horizontal orientation. Furthermore, the effectiveness of our methods to detect text strings with arbitrary orientations is evaluated on the Oriented Scene Text Dataset collected by ourselves containing text strings in non-horizontal orientations.
Url:
DOI: 10.1109/TIP.2011.2126586
PubMed: 21411405
PubMed Central: 3337634
Links toward previous steps (curation, corpus...)
- to stream Pmc, to step Corpus: Pour aller vers cette notice dans l'étape Curation :000145
Links to Exploration step
PMC:3337634Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Text String Detection from Natural Scenes by Structure-based Partition and Grouping</title>
<author><name sortKey="Yi, Chucai" sort="Yi, Chucai" uniqKey="Yi C" first="Chucai" last="Yi">Chucai Yi</name>
</author>
<author><name sortKey="Tian, Yingli" sort="Tian, Yingli" uniqKey="Tian Y" first="Yingli" last="Tian">Yingli Tian</name>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">PMC</idno>
<idno type="pmid">21411405</idno>
<idno type="pmc">3337634</idno>
<idno type="url">http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3337634</idno>
<idno type="RBID">PMC:3337634</idno>
<idno type="doi">10.1109/TIP.2011.2126586</idno>
<date when="2011">2011</date>
<idno type="wicri:Area/Pmc/Corpus">000145</idno>
<idno type="wicri:Area/Pmc/Curation">000145</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a" type="main">Text String Detection from Natural Scenes by Structure-based Partition and Grouping</title>
<author><name sortKey="Yi, Chucai" sort="Yi, Chucai" uniqKey="Yi C" first="Chucai" last="Yi">Chucai Yi</name>
</author>
<author><name sortKey="Tian, Yingli" sort="Tian, Yingli" uniqKey="Tian Y" first="Yingli" last="Tian">Yingli Tian</name>
</author>
</analytic>
<series><title level="j">Ieee Transactions on Image Processing</title>
<idno type="ISSN">1057-7149</idno>
<idno type="eISSN">1941-0042</idno>
<imprint><date when="2011">2011</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
</fileDesc>
<profileDesc><textClass></textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en"><p id="P1">Text information in natural scene images serves as important clues for many image-based applications such as scene understanding, content-based image retrieval, assistive navigation, and automatic geocoding. However, locating text from complex background with multiple colors is a challenging task. In this paper, we explore a new framework to detect text strings with arbitrary orientations in complex natural scene images. Our proposed framework of text string detection consists of two steps: 1) Image partition to find text character candidates based on local gradient features and color uniformity of character components. 2) Character candidate grouping to detect text strings based on joint structural features of text characters in each text string such as character size differences, distances between neighboring characters, and character alignment. By assuming that a text string has at least three characters, we propose two algorithms of text string detection: 1) adjacent character grouping method, and 2) text line grouping method. The adjacent character grouping method calculates the sibling groups of each character candidate as string segments and then merges the intersecting sibling groups into text string. The text line grouping method performs Hough transform to fit text line among the centroids of text candidates. Each fitted text line describes the orientation of a potential text string. The detected text string is presented by a rectangle region covering all characters whose centroids are cascaded in its text line. To improve efficiency and accuracy, our algorithms are carried out in multi-scales. The proposed methods outperform the state-of-the-art results on the public Robust Reading Dataset which contains text only in horizontal orientation. Furthermore, the effectiveness of our methods to detect text strings with arbitrary orientations is evaluated on the Oriented Scene Text Dataset collected by ourselves containing text strings in non-horizontal orientations.</p>
</div>
</front>
</TEI>
<pmc article-type="research-article"><pmc-comment>The publisher of this article does not allow downloading of the full text in XML form.</pmc-comment>
<pmc-dir>properties manuscript</pmc-dir>
<front><journal-meta><journal-id journal-id-type="nlm-journal-id">9886191</journal-id>
<journal-id journal-id-type="pubmed-jr-id">22899</journal-id>
<journal-id journal-id-type="nlm-ta">IEEE Trans Image Process</journal-id>
<journal-id journal-id-type="iso-abbrev">IEEE Trans Image Process</journal-id>
<journal-title-group><journal-title>Ieee Transactions on Image Processing</journal-title>
</journal-title-group>
<issn pub-type="ppub">1057-7149</issn>
<issn pub-type="epub">1941-0042</issn>
</journal-meta>
<article-meta><article-id pub-id-type="pmid">21411405</article-id>
<article-id pub-id-type="pmc">3337634</article-id>
<article-id pub-id-type="doi">10.1109/TIP.2011.2126586</article-id>
<article-id pub-id-type="manuscript">NIHMS369669</article-id>
<article-categories><subj-group subj-group-type="heading"><subject>Article</subject>
</subj-group>
</article-categories>
<title-group><article-title>Text String Detection from Natural Scenes by Structure-based Partition and Grouping</article-title>
</title-group>
<contrib-group><contrib contrib-type="author"><name><surname>Yi</surname>
<given-names>Chucai</given-names>
</name>
<aff id="A1">Graduate Center, City University of New York, New York, NY 10016 USA (phone: 212-650-8917; fax: 212-650-8249;<email>cyi@gc.cuny.edu</email>
)</aff>
</contrib>
<contrib contrib-type="author"><name><surname>Tian</surname>
<given-names>YingLi</given-names>
</name>
<role>Senior Member, IEEE</role>
<aff id="A2">City College, City University of New York, New York, NY 10031 USA (<email>ytian@ccny.cuny.edu</email>
). Prior to joining the City College in September 2008, she was with IBM T.J. Watson Research Center, Yorktown Heights, NY 10598 USA</aff>
</contrib>
</contrib-group>
<pub-date pub-type="nihms-submitted"><day>11</day>
<month>4</month>
<year>2012</year>
</pub-date>
<pub-date pub-type="epub"><day>14</day>
<month>3</month>
<year>2011</year>
</pub-date>
<pub-date pub-type="ppub"><month>9</month>
<year>2011</year>
</pub-date>
<pub-date pub-type="pmc-release"><day>26</day>
<month>4</month>
<year>2012</year>
</pub-date>
<volume>20</volume>
<issue>9</issue>
<fpage>2594</fpage>
<lpage>2605</lpage>
<permissions><copyright-statement>Copyright © 2011 IEEE.</copyright-statement>
<copyright-year>2011</copyright-year>
</permissions>
<abstract><p id="P1">Text information in natural scene images serves as important clues for many image-based applications such as scene understanding, content-based image retrieval, assistive navigation, and automatic geocoding. However, locating text from complex background with multiple colors is a challenging task. In this paper, we explore a new framework to detect text strings with arbitrary orientations in complex natural scene images. Our proposed framework of text string detection consists of two steps: 1) Image partition to find text character candidates based on local gradient features and color uniformity of character components. 2) Character candidate grouping to detect text strings based on joint structural features of text characters in each text string such as character size differences, distances between neighboring characters, and character alignment. By assuming that a text string has at least three characters, we propose two algorithms of text string detection: 1) adjacent character grouping method, and 2) text line grouping method. The adjacent character grouping method calculates the sibling groups of each character candidate as string segments and then merges the intersecting sibling groups into text string. The text line grouping method performs Hough transform to fit text line among the centroids of text candidates. Each fitted text line describes the orientation of a potential text string. The detected text string is presented by a rectangle region covering all characters whose centroids are cascaded in its text line. To improve efficiency and accuracy, our algorithms are carried out in multi-scales. The proposed methods outperform the state-of-the-art results on the public Robust Reading Dataset which contains text only in horizontal orientation. Furthermore, the effectiveness of our methods to detect text strings with arbitrary orientations is evaluated on the Oriented Scene Text Dataset collected by ourselves containing text strings in non-horizontal orientations.</p>
</abstract>
<kwd-group><title>Index Terms</title>
<kwd>Adjacent character grouping</kwd>
<kwd>Character property</kwd>
<kwd>Image partition</kwd>
<kwd>Text line grouping</kwd>
<kwd>Text string detection</kwd>
<kwd>Text string structure</kwd>
</kwd-group>
<funding-group><award-group><funding-source country="United States">National Eye Institute : NEI</funding-source>
<award-id>R21 EY020990-01 || EY</award-id>
</award-group>
</funding-group>
</article-meta>
</front>
</pmc>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Pmc/Curation
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000145 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Pmc/Curation/biblio.hfd -nk 000145 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Ticri/CIDE |area= OcrV1 |flux= Pmc |étape= Curation |type= RBID |clé= PMC:3337634 |texte= Text String Detection from Natural Scenes by Structure-based Partition and Grouping }}
Pour générer des pages wiki
HfdIndexSelect -h $EXPLOR_AREA/Data/Pmc/Curation/RBID.i -Sk "pubmed:21411405" \ | HfdSelect -Kh $EXPLOR_AREA/Data/Pmc/Curation/biblio.hfd \ | NlmPubMed2Wicri -a OcrV1
This area was generated with Dilib version V0.6.32. |