Text Extraction from Scene Images by Character Appearance and Structure Modeling
Identifieur interne : 000158 ( Main/Merge ); précédent : 000157; suivant : 000159Text Extraction from Scene Images by Character Appearance and Structure Modeling
Auteurs : Chucai Yi ; Yingli TianSource :
- Computer vision and image understanding : CVIU [ 1077-3142 ] ; 2013.
Abstract
In this paper, we propose a novel algorithm to detect text information from natural scene images. Scene text classification and detection are still open research topics. Our proposed algorithm is able to model both character appearance and structure to generate representative and discriminative text descriptors. The contributions of this paper include three aspects: 1) a new character appearance model by a structure correlation algorithm which extracts discriminative appearance features from detected interest points of character samples; 2) a new text descriptor based on structons and correlatons, which model character structure by structure differences among character samples and structure component co-occurrence; and 3) a new text region localization method by combining color decomposition, character contour refinement, and string line alignment to localize character candidates and refine detected text regions. We perform three groups of experiments to evaluate the effectiveness of our proposed algorithm, including text classification, text detection, and character identification. The evaluation results on benchmark datasets demonstrate that our algorithm achieves the state-of-the-art performance on scene text classification and detection, and significantly outperforms the existing algorithms for character identification.
Url:
DOI: 10.1016/j.cviu.2012.11.002
PubMed: 23316111
PubMed Central: 3539806
Links toward previous steps (curation, corpus...)
- to stream Pmc, to step Corpus: 000147
- to stream Pmc, to step Curation: 000147
- to stream Pmc, to step Checkpoint: 000065
- to stream Ncbi, to step Merge: 000153
- to stream Ncbi, to step Curation: 000153
- to stream Ncbi, to step Checkpoint: 000153
Links to Exploration step
PMC:3539806Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Text Extraction from Scene Images by Character Appearance and Structure Modeling</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">23316111</idno>
<idno type="pmc">3539806</idno>
<idno type="url">http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3539806</idno>
<idno type="RBID">PMC:3539806</idno>
<idno type="doi">10.1016/j.cviu.2012.11.002</idno>
<date when="2013">2013</date>
<idno type="wicri:Area/Pmc/Corpus">000147</idno>
<idno type="wicri:Area/Pmc/Curation">000147</idno>
<idno type="wicri:Area/Pmc/Checkpoint">000065</idno>
<idno type="wicri:Area/Ncbi/Merge">000153</idno>
<idno type="wicri:Area/Ncbi/Curation">000153</idno>
<idno type="wicri:Area/Ncbi/Checkpoint">000153</idno>
<idno type="wicri:doubleKey">1077-3142:2013:Yi C:text:extraction:from</idno>
<idno type="wicri:Area/Main/Merge">000158</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a" type="main">Text Extraction from Scene Images by Character Appearance and Structure Modeling</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">Computer vision and image understanding : CVIU</title>
<idno type="ISSN">1077-3142</idno>
<imprint><date when="2013">2013</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
</fileDesc>
<profileDesc><textClass></textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en"><p id="P2">In this paper, we propose a novel algorithm to detect text information from natural scene images. Scene text classification and detection are still open research topics. Our proposed algorithm is able to model both character appearance and structure to generate representative and discriminative text descriptors. The contributions of this paper include three aspects: 1) a new character appearance model by a structure correlation algorithm which extracts discriminative appearance features from detected interest points of character samples; 2) a new text descriptor based on structons and correlatons, which model character structure by structure differences among character samples and structure component co-occurrence; and 3) a new text region localization method by combining color decomposition, character contour refinement, and string line alignment to localize character candidates and refine detected text regions. We perform three groups of experiments to evaluate the effectiveness of our proposed algorithm, including text classification, text detection, and character identification. The evaluation results on benchmark datasets demonstrate that our algorithm achieves the state-of-the-art performance on scene text classification and detection, and significantly outperforms the existing algorithms for character identification.</p>
</div>
</front>
</TEI>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Main/Merge
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000158 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Merge/biblio.hfd -nk 000158 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Ticri/CIDE |area= OcrV1 |flux= Main |étape= Merge |type= RBID |clé= PMC:3539806 |texte= Text Extraction from Scene Images by Character Appearance and Structure Modeling }}
Pour générer des pages wiki
HfdIndexSelect -h $EXPLOR_AREA/Data/Main/Merge/RBID.i -Sk "pubmed:23316111" \ | HfdSelect -Kh $EXPLOR_AREA/Data/Main/Merge/biblio.hfd \ | NlmPubMed2Wicri -a OcrV1
This area was generated with Dilib version V0.6.32. |