Adaptive color document images binarization for text retrieval
Identifieur interne : 001654 ( Main/Exploration ); précédent : 001653; suivant : 001655Adaptive color document images binarization for text retrieval
Auteurs : YI LI [République populaire de Chine] ; ZHIYAN WANG [République populaire de Chine] ; HAIZAN ZENG [République populaire de Chine]Source :
- SPIE proceedings series [ 1017-2653 ] ; 2004.
Descripteurs français
- Pascal (Inist)
- Traitement image, Reconnaissance forme, Image couleur, Recherche documentaire, Recherche image, Recherche information, Texte, Arbre décision, Reconnaissance caractère, Reconnaissance optique caractère, Détection seuil, Saturation, Méthode adaptative, Ecart type, Transformation Karhunen Loeve, Dimension corrélation, Loi normale.
- Wicri :
- topic : Recherche documentaire.
English descriptors
- KwdEn :
- Adaptive method, Character recognition, Color image, Correlation dimension, Decision tree, Document retrieval, Gaussian distribution, Image processing, Image retrieval, Information retrieval, Karhunen Loeve transformation, Optical character recognition, Pattern recognition, Saturation, Standard deviation, Text, Threshold detection.
Abstract
This paper presents a decision tree based adaptive binarization method for text retrieval in color document images. This method extends Ni-Black windowed thresholding technique and hue (H), saturation (S) and value (V) are employed. First, an observation window is retrieved, and based on standard deviation of H, S and V, a pre-defined decision tree is used for selecting proper variables that should be employed. Secondly, Karhunen-Loeve Transform (KLT) is used for eliminating correlation and reducing dimension. Finally, center point of the window is classified based on 2-D standard normal distribution. The result shows that our binarization method generates better result than Ni-Black and other global thresholding binarization method such as Otsu's in color document images. A comparison using a commercial OCR system shows that our method can be used in various situations for high quality text retrieval.
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream PascalFrancis, to step Corpus: 000515
- to stream PascalFrancis, to step Curation: 000274
- to stream PascalFrancis, to step Checkpoint: 000513
- to stream Main, to step Merge: 001724
- to stream Main, to step Curation: 001654
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">Adaptive color document images binarization for text retrieval</title>
<author><name sortKey="Yi Li" sort="Yi Li" uniqKey="Yi Li" last="Yi Li">YI LI</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>School of Computer Science and Engineering, South China University of Technology</s1>
<s2>Wushan, Guangzhou, 510640</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
<wicri:noRegion>Wushan, Guangzhou, 510640</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Zhiyan Wang" sort="Zhiyan Wang" uniqKey="Zhiyan Wang" last="Zhiyan Wang">ZHIYAN WANG</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>School of Computer Science and Engineering, South China University of Technology</s1>
<s2>Wushan, Guangzhou, 510640</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
<wicri:noRegion>Wushan, Guangzhou, 510640</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Haizan Zeng" sort="Haizan Zeng" uniqKey="Haizan Zeng" last="Haizan Zeng">HAIZAN ZENG</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>School of Computer Science and Engineering, South China University of Technology</s1>
<s2>Wushan, Guangzhou, 510640</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
<wicri:noRegion>Wushan, Guangzhou, 510640</wicri:noRegion>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">INIST</idno>
<idno type="inist">04-0535107</idno>
<date when="2004">2004</date>
<idno type="stanalyst">PASCAL 04-0535107 INIST</idno>
<idno type="RBID">Pascal:04-0535107</idno>
<idno type="wicri:Area/PascalFrancis/Corpus">000515</idno>
<idno type="wicri:Area/PascalFrancis/Curation">000274</idno>
<idno type="wicri:Area/PascalFrancis/Checkpoint">000513</idno>
<idno type="wicri:doubleKey">1017-2653:2004:Yi Li:adaptive:color:document</idno>
<idno type="wicri:Area/Main/Merge">001724</idno>
<idno type="wicri:Area/Main/Curation">001654</idno>
<idno type="wicri:Area/Main/Exploration">001654</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">Adaptive color document images binarization for text retrieval</title>
<author><name sortKey="Yi Li" sort="Yi Li" uniqKey="Yi Li" last="Yi Li">YI LI</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>School of Computer Science and Engineering, South China University of Technology</s1>
<s2>Wushan, Guangzhou, 510640</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
<wicri:noRegion>Wushan, Guangzhou, 510640</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Zhiyan Wang" sort="Zhiyan Wang" uniqKey="Zhiyan Wang" last="Zhiyan Wang">ZHIYAN WANG</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>School of Computer Science and Engineering, South China University of Technology</s1>
<s2>Wushan, Guangzhou, 510640</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
<wicri:noRegion>Wushan, Guangzhou, 510640</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Haizan Zeng" sort="Haizan Zeng" uniqKey="Haizan Zeng" last="Haizan Zeng">HAIZAN ZENG</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>School of Computer Science and Engineering, South China University of Technology</s1>
<s2>Wushan, Guangzhou, 510640</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
<wicri:noRegion>Wushan, Guangzhou, 510640</wicri:noRegion>
</affiliation>
</author>
</analytic>
<series><title level="j" type="main">SPIE proceedings series</title>
<idno type="ISSN">1017-2653</idno>
<imprint><date when="2004">2004</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><title level="j" type="main">SPIE proceedings series</title>
<idno type="ISSN">1017-2653</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Adaptive method</term>
<term>Character recognition</term>
<term>Color image</term>
<term>Correlation dimension</term>
<term>Decision tree</term>
<term>Document retrieval</term>
<term>Gaussian distribution</term>
<term>Image processing</term>
<term>Image retrieval</term>
<term>Information retrieval</term>
<term>Karhunen Loeve transformation</term>
<term>Optical character recognition</term>
<term>Pattern recognition</term>
<term>Saturation</term>
<term>Standard deviation</term>
<term>Text</term>
<term>Threshold detection</term>
</keywords>
<keywords scheme="Pascal" xml:lang="fr"><term>Traitement image</term>
<term>Reconnaissance forme</term>
<term>Image couleur</term>
<term>Recherche documentaire</term>
<term>Recherche image</term>
<term>Recherche information</term>
<term>Texte</term>
<term>Arbre décision</term>
<term>Reconnaissance caractère</term>
<term>Reconnaissance optique caractère</term>
<term>Détection seuil</term>
<term>Saturation</term>
<term>Méthode adaptative</term>
<term>Ecart type</term>
<term>Transformation Karhunen Loeve</term>
<term>Dimension corrélation</term>
<term>Loi normale</term>
</keywords>
<keywords scheme="Wicri" type="topic" xml:lang="fr"><term>Recherche documentaire</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">This paper presents a decision tree based adaptive binarization method for text retrieval in color document images. This method extends Ni-Black windowed thresholding technique and hue (H), saturation (S) and value (V) are employed. First, an observation window is retrieved, and based on standard deviation of H, S and V, a pre-defined decision tree is used for selecting proper variables that should be employed. Secondly, Karhunen-Loeve Transform (KLT) is used for eliminating correlation and reducing dimension. Finally, center point of the window is classified based on 2-D standard normal distribution. The result shows that our binarization method generates better result than Ni-Black and other global thresholding binarization method such as Otsu's in color document images. A comparison using a commercial OCR system shows that our method can be used in various situations for high quality text retrieval.</div>
</front>
</TEI>
<affiliations><list><country><li>République populaire de Chine</li>
</country>
</list>
<tree><country name="République populaire de Chine"><noRegion><name sortKey="Yi Li" sort="Yi Li" uniqKey="Yi Li" last="Yi Li">YI LI</name>
</noRegion>
<name sortKey="Haizan Zeng" sort="Haizan Zeng" uniqKey="Haizan Zeng" last="Haizan Zeng">HAIZAN ZENG</name>
<name sortKey="Zhiyan Wang" sort="Zhiyan Wang" uniqKey="Zhiyan Wang" last="Zhiyan Wang">ZHIYAN WANG</name>
</country>
</tree>
</affiliations>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 001654 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 001654 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Ticri/CIDE |area= OcrV1 |flux= Main |étape= Exploration |type= RBID |clé= Pascal:04-0535107 |texte= Adaptive color document images binarization for text retrieval }}
This area was generated with Dilib version V0.6.32. |