Handwritten Gujarati Numeral Optical Character Recognition using Hybrid Feature Extraction Technique
Identifieur interne : 000778 ( Main/Exploration ); précédent : 000777; suivant : 000779Handwritten Gujarati Numeral Optical Character Recognition using Hybrid Feature Extraction Technique
Auteurs : Apurva A. Desai [Inde]Source :
Descripteurs français
- Pascal (Inist)
- Wicri :
- topic : Classification.
English descriptors
- KwdEn :
Abstract
Lot of work has been done for Optical Character Recognition (OCR) for various Indian languages. But for Gujarati, a language belonging to Devnagari family of languages and spoken in the western state of Gujarat in India, hardly any work can be traced especially for handwritten characters. In this work I have addressed the problem of handwritten Gujarati numerals. Here pre-process techniques like global threshold, erosion and dilation, skew correction etc, are used. A novel hybrid feature extraction technique is suggested and used which is constituted by a structural approach and statistical approach of feature extraction. Image is subdivided and then the pixel information is used as a structural approach whereas the aspect ratio of the number is considered as a statistical approach. For classification kNN classifier has been used. This model gives overall accuracy of 96.99% for the handwritten Gujarati numerals.
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream PascalFrancis, to step Corpus: 000089
- to stream PascalFrancis, to step Curation: 000683
- to stream PascalFrancis, to step Checkpoint: 000152
- to stream Main, to step Merge: 000783
- to stream Main, to step Curation: 000778
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">Handwritten Gujarati Numeral Optical Character Recognition using Hybrid Feature Extraction Technique</title><author><name sortKey="Desai, Apurva A" sort="Desai, Apurva A" uniqKey="Desai A" first="Apurva A." last="Desai">Apurva A. Desai</name><affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Department of Computer Science, Veer Narmad South Gujarat University, Udhna Magdalla Road</s1><s2>Surat - 395007, Gujarat</s2><s3>IND</s3><sZ>1 aut.</sZ></inist:fA14><country>Inde</country><wicri:noRegion>Surat - 395007, Gujarat</wicri:noRegion></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">INIST</idno><idno type="inist">12-0306480</idno><date when="2010">2010</date><idno type="stanalyst">PASCAL 12-0306480 INIST</idno><idno type="RBID">Pascal:12-0306480</idno><idno type="wicri:Area/PascalFrancis/Corpus">000089</idno><idno type="wicri:Area/PascalFrancis/Curation">000683</idno><idno type="wicri:Area/PascalFrancis/Checkpoint">000152</idno><idno type="wicri:Area/Main/Merge">000783</idno><idno type="wicri:Area/Main/Curation">000778</idno><idno type="wicri:Area/Main/Exploration">000778</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">Handwritten Gujarati Numeral Optical Character Recognition using Hybrid Feature Extraction Technique</title><author><name sortKey="Desai, Apurva A" sort="Desai, Apurva A" uniqKey="Desai A" first="Apurva A." last="Desai">Apurva A. Desai</name><affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Department of Computer Science, Veer Narmad South Gujarat University, Udhna Magdalla Road</s1><s2>Surat - 395007, Gujarat</s2><s3>IND</s3><sZ>1 aut.</sZ></inist:fA14><country>Inde</country><wicri:noRegion>Surat - 395007, Gujarat</wicri:noRegion></affiliation></author></analytic></biblStruct></sourceDesc></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Aspect ratio</term><term>Character recognition</term><term>Classification</term><term>Feature extraction</term><term>Language family</term><term>Manuscript character</term><term>Mathematical morphology</term><term>Modeling</term><term>Natural language</term><term>Optical character recognition</term><term>Pattern extraction</term><term>Pattern recognition</term><term>Statistical analysis</term></keywords><keywords scheme="Pascal" xml:lang="fr"><term>Reconnaissance caractère</term><term>Reconnaissance forme</term><term>Caractère manuscrit</term><term>Reconnaissance optique caractère</term><term>Famille langage</term><term>Langage naturel</term><term>Morphologie mathématique</term><term>Classification</term><term>Extraction caractéristique</term><term>Rapport aspect</term><term>Extraction forme</term><term>Analyse statistique</term><term>Modélisation</term><term>.</term></keywords><keywords scheme="Wicri" type="topic" xml:lang="fr"><term>Classification</term></keywords></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Lot of work has been done for Optical Character Recognition (OCR) for various Indian languages. But for Gujarati, a language belonging to Devnagari family of languages and spoken in the western state of Gujarat in India, hardly any work can be traced especially for handwritten characters. In this work I have addressed the problem of handwritten Gujarati numerals. Here pre-process techniques like global threshold, erosion and dilation, skew correction etc, are used. A novel hybrid feature extraction technique is suggested and used which is constituted by a structural approach and statistical approach of feature extraction. Image is subdivided and then the pixel information is used as a structural approach whereas the aspect ratio of the number is considered as a statistical approach. For classification kNN classifier has been used. This model gives overall accuracy of 96.99% for the handwritten Gujarati numerals.</div></front></TEI><affiliations><list><country><li>Inde</li></country></list><tree><country name="Inde"><noRegion><name sortKey="Desai, Apurva A" sort="Desai, Apurva A" uniqKey="Desai A" first="Apurva A." last="Desai">Apurva A. Desai</name></noRegion></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 000778 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000778 | 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:12-0306480
|texte= Handwritten Gujarati Numeral Optical Character Recognition using Hybrid Feature Extraction Technique
}}
| This area was generated with Dilib version V0.6.32. |  |