Real-time Lexicon-free Scene Text Localization and Recognition.
Identifieur interne : 000243 ( Ncbi/Merge ); précédent : 000242; suivant : 000244Real-time Lexicon-free Scene Text Localization and Recognition.
Auteurs : Lukas Neumann ; Jiri MatasSource :
- IEEE transactions on pattern analysis and machine intelligence [ 1939-3539 ] ; 2015.
Abstract
An end-to-end real-time text localization and recognition method is presented. Its real-time performance is achieved by posing the character detection and segmentation problem as an efficient sequential selection from the set of Extremal Regions. The ER detector is robust against blur, low contrast and illumination, color and texture variation. In the first stage, the probability of each ER being a character is estimated using features calculated by a novel algorithm in constant time and only ERs with locally maximal probability are selected for the second stage, where the classification accuracy is improved using computationally more expensive features. A highly efficient clustering algorithm then groups ERs into text lines and an OCR classifier trained on synthetic fonts is exploited to label character regions. The most probable character sequence is selected in the last stage when the context of each character is known. The method was evaluated on three public datasets. On the ICDAR 2013 dataset the method achieves state-of-the-art results in text localization; on the more challenging SVT dataset, the proposed method significantly outperforms the state-of-the-art methods and demonstrates that the proposed pipeline can incorporate additional prior knowledge about the detected text. The proposed method was exploited as the baseline in the ICDAR 2015 Robust Reading competition, where it compares favourably to the state-of-the art.
DOI: 10.1109/TPAMI.2015.2496234
PubMed: 26540676
Links toward previous steps (curation, corpus...)
- to stream PubMed, to step Corpus: 000004
- to stream PubMed, to step Curation: 000004
- to stream PubMed, to step Checkpoint: 000004
Links to Exploration step
pubmed:26540676Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Real-time Lexicon-free Scene Text Localization and Recognition.</title>
<author><name sortKey="Neumann, Lukas" sort="Neumann, Lukas" uniqKey="Neumann L" first="Lukas" last="Neumann">Lukas Neumann</name>
</author>
<author><name sortKey="Matas, Jiri" sort="Matas, Jiri" uniqKey="Matas J" first="Jiri" last="Matas">Jiri Matas</name>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">PubMed</idno>
<date when="2015">2015</date>
<idno type="doi">10.1109/TPAMI.2015.2496234</idno>
<idno type="RBID">pubmed:26540676</idno>
<idno type="pmid">26540676</idno>
<idno type="wicri:Area/PubMed/Corpus">000004</idno>
<idno type="wicri:Area/PubMed/Curation">000004</idno>
<idno type="wicri:Area/PubMed/Checkpoint">000004</idno>
<idno type="wicri:Area/Ncbi/Merge">000243</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en">Real-time Lexicon-free Scene Text Localization and Recognition.</title>
<author><name sortKey="Neumann, Lukas" sort="Neumann, Lukas" uniqKey="Neumann L" first="Lukas" last="Neumann">Lukas Neumann</name>
</author>
<author><name sortKey="Matas, Jiri" sort="Matas, Jiri" uniqKey="Matas J" first="Jiri" last="Matas">Jiri Matas</name>
</author>
</analytic>
<series><title level="j">IEEE transactions on pattern analysis and machine intelligence</title>
<idno type="eISSN">1939-3539</idno>
<imprint><date when="2015" type="published">2015</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
</fileDesc>
<profileDesc><textClass></textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">An end-to-end real-time text localization and recognition method is presented. Its real-time performance is achieved by posing the character detection and segmentation problem as an efficient sequential selection from the set of Extremal Regions. The ER detector is robust against blur, low contrast and illumination, color and texture variation. In the first stage, the probability of each ER being a character is estimated using features calculated by a novel algorithm in constant time and only ERs with locally maximal probability are selected for the second stage, where the classification accuracy is improved using computationally more expensive features. A highly efficient clustering algorithm then groups ERs into text lines and an OCR classifier trained on synthetic fonts is exploited to label character regions. The most probable character sequence is selected in the last stage when the context of each character is known. The method was evaluated on three public datasets. On the ICDAR 2013 dataset the method achieves state-of-the-art results in text localization; on the more challenging SVT dataset, the proposed method significantly outperforms the state-of-the-art methods and demonstrates that the proposed pipeline can incorporate additional prior knowledge about the detected text. The proposed method was exploited as the baseline in the ICDAR 2015 Robust Reading competition, where it compares favourably to the state-of-the art.</div>
</front>
</TEI>
<pubmed><MedlineCitation Status="Publisher" Owner="NLM"><PMID Version="1">26540676</PMID>
<DateCreated><Year>2015</Year>
<Month>11</Month>
<Day>5</Day>
</DateCreated>
<DateRevised><Year>2015</Year>
<Month>11</Month>
<Day>6</Day>
</DateRevised>
<Article PubModel="Print-Electronic"><Journal><ISSN IssnType="Electronic">1939-3539</ISSN>
<JournalIssue CitedMedium="Internet"><PubDate><Year>2015</Year>
<Month>Oct</Month>
<Day>30</Day>
</PubDate>
</JournalIssue>
<Title>IEEE transactions on pattern analysis and machine intelligence</Title>
<ISOAbbreviation>IEEE Trans Pattern Anal Mach Intell</ISOAbbreviation>
</Journal>
<ArticleTitle>Real-time Lexicon-free Scene Text Localization and Recognition.</ArticleTitle>
<Pagination><MedlinePgn></MedlinePgn>
</Pagination>
<Abstract><AbstractText NlmCategory="UNASSIGNED">An end-to-end real-time text localization and recognition method is presented. Its real-time performance is achieved by posing the character detection and segmentation problem as an efficient sequential selection from the set of Extremal Regions. The ER detector is robust against blur, low contrast and illumination, color and texture variation. In the first stage, the probability of each ER being a character is estimated using features calculated by a novel algorithm in constant time and only ERs with locally maximal probability are selected for the second stage, where the classification accuracy is improved using computationally more expensive features. A highly efficient clustering algorithm then groups ERs into text lines and an OCR classifier trained on synthetic fonts is exploited to label character regions. The most probable character sequence is selected in the last stage when the context of each character is known. The method was evaluated on three public datasets. On the ICDAR 2013 dataset the method achieves state-of-the-art results in text localization; on the more challenging SVT dataset, the proposed method significantly outperforms the state-of-the-art methods and demonstrates that the proposed pipeline can incorporate additional prior knowledge about the detected text. The proposed method was exploited as the baseline in the ICDAR 2015 Robust Reading competition, where it compares favourably to the state-of-the art.</AbstractText>
</Abstract>
<AuthorList><Author><LastName>Neumann</LastName>
<ForeName>Lukas</ForeName>
<Initials>L</Initials>
</Author>
<Author><LastName>Matas</LastName>
<ForeName>Jiri</ForeName>
<Initials>J</Initials>
</Author>
</AuthorList>
<Language>ENG</Language>
<PublicationTypeList><PublicationType UI="">JOURNAL ARTICLE</PublicationType>
</PublicationTypeList>
<ArticleDate DateType="Electronic"><Year>2015</Year>
<Month>10</Month>
<Day>30</Day>
</ArticleDate>
</Article>
<MedlineJournalInfo><MedlineTA>IEEE Trans Pattern Anal Mach Intell</MedlineTA>
<NlmUniqueID>9885960</NlmUniqueID>
<ISSNLinking>0098-5589</ISSNLinking>
</MedlineJournalInfo>
</MedlineCitation>
<PubmedData><History><PubMedPubDate PubStatus="entrez"><Year>2015</Year>
<Month>11</Month>
<Day>6</Day>
<Hour>6</Hour>
<Minute>0</Minute>
</PubMedPubDate>
<PubMedPubDate PubStatus="pubmed"><Year>2015</Year>
<Month>11</Month>
<Day>6</Day>
<Hour>6</Hour>
<Minute>0</Minute>
</PubMedPubDate>
<PubMedPubDate PubStatus="medline"><Year>2015</Year>
<Month>11</Month>
<Day>6</Day>
<Hour>6</Hour>
<Minute>0</Minute>
</PubMedPubDate>
</History>
<PublicationStatus>aheadofprint</PublicationStatus>
<ArticleIdList><ArticleId IdType="doi">10.1109/TPAMI.2015.2496234</ArticleId>
<ArticleId IdType="pubmed">26540676</ArticleId>
</ArticleIdList>
</PubmedData>
</pubmed>
<affiliations><list></list>
<tree><noCountry><name sortKey="Matas, Jiri" sort="Matas, Jiri" uniqKey="Matas J" first="Jiri" last="Matas">Jiri Matas</name>
<name sortKey="Neumann, Lukas" sort="Neumann, Lukas" uniqKey="Neumann L" first="Lukas" last="Neumann">Lukas Neumann</name>
</noCountry>
</tree>
</affiliations>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Ncbi/Merge
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000243 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Ncbi/Merge/biblio.hfd -nk 000243 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Ticri/CIDE |area= OcrV1 |flux= Ncbi |étape= Merge |type= RBID |clé= pubmed:26540676 |texte= Real-time Lexicon-free Scene Text Localization and Recognition. }}
Pour générer des pages wiki
HfdIndexSelect -h $EXPLOR_AREA/Data/Ncbi/Merge/RBID.i -Sk "pubmed:26540676" \ | HfdSelect -Kh $EXPLOR_AREA/Data/Ncbi/Merge/biblio.hfd \ | NlmPubMed2Wicri -a OcrV1
This area was generated with Dilib version V0.6.32. |