New Approach of Feature Extraction Method Based on the Raw Form and his Skeleton for Gujarati Handwritten Digits using Neural Networks Classifier

This paper presents an optical character recognition (OCR) system for Gujarati handwritten digits. One may find so much of work for latin writing, arabic, chines, etc. but Gujarati is a language for which hardly any work is traceable especially for handwritten characters. Here in this work we have p...

Full description

Bibliographic Details
Main Authors: K. Moro, Mohammed Fakir, Belaid Bouikhalene, Rachid El Yachi, Bader Dinne El Kessab
Format: Article
Language:English
Published: Universitatea Dunarea de Jos 2014-12-01
Series:Analele Universităţii "Dunărea de Jos" Galaţi: Fascicula III, Electrotehnică, Electronică, Automatică, Informatică
Subjects:
Online Access:http://www.ann.ugal.ro/eeai/archives/2014/1%20NEW%20APPROACH%20OF%20FEATURE%20EXTRACTION%20METHOD.pdf
id doaj-a6b7c8b9eea44ffe8c04188f5c2b98ec
record_format Article
spelling doaj-a6b7c8b9eea44ffe8c04188f5c2b98ec2020-11-24T22:28:54ZengUniversitatea Dunarea de JosAnalele Universităţii "Dunărea de Jos" Galaţi: Fascicula III, Electrotehnică, Electronică, Automatică, Informatică1221-454X1221-454X2014-12-01371510New Approach of Feature Extraction Method Based on the Raw Form and his Skeleton for Gujarati Handwritten Digits using Neural Networks ClassifierK. Moro0Mohammed Fakir1Belaid Bouikhalene2Rachid El Yachi3Bader Dinne El Kessab4Faculty of Science and Technology Sultan Moulay Slimane University 523, Beni Mellal, MoroccoFaculty of Science and Technology Sultan Moulay Slimane University 523, Beni Mellal, MoroccoFaculty of Science and Technology Sultan Moulay Slimane University 523, Beni Mellal, MoroccoFaculty of Science and Technology Sultan Moulay Slimane University 523, Beni Mellal, MoroccoFaculty of Science and Technology Sultan Moulay Slimane University 523, Beni Mellal, MoroccoThis paper presents an optical character recognition (OCR) system for Gujarati handwritten digits. One may find so much of work for latin writing, arabic, chines, etc. but Gujarati is a language for which hardly any work is traceable especially for handwritten characters. Here in this work we have proposed a method of feature extraction based on the raw form of the character and his skeleton and we have shown the advantage of using this method over other approaches mentioned in this article.http://www.ann.ugal.ro/eeai/archives/2014/1%20NEW%20APPROACH%20OF%20FEATURE%20EXTRACTION%20METHOD.pdfOptical character recognitionneural networkfeature extractionGujarati handwritten digitsskeletonizationclassification
collection DOAJ
language English
format Article
sources DOAJ
author K. Moro
Mohammed Fakir
Belaid Bouikhalene
Rachid El Yachi
Bader Dinne El Kessab
spellingShingle K. Moro
Mohammed Fakir
Belaid Bouikhalene
Rachid El Yachi
Bader Dinne El Kessab
New Approach of Feature Extraction Method Based on the Raw Form and his Skeleton for Gujarati Handwritten Digits using Neural Networks Classifier
Analele Universităţii "Dunărea de Jos" Galaţi: Fascicula III, Electrotehnică, Electronică, Automatică, Informatică
Optical character recognition
neural network
feature extraction
Gujarati handwritten digits
skeletonization
classification
author_facet K. Moro
Mohammed Fakir
Belaid Bouikhalene
Rachid El Yachi
Bader Dinne El Kessab
author_sort K. Moro
title New Approach of Feature Extraction Method Based on the Raw Form and his Skeleton for Gujarati Handwritten Digits using Neural Networks Classifier
title_short New Approach of Feature Extraction Method Based on the Raw Form and his Skeleton for Gujarati Handwritten Digits using Neural Networks Classifier
title_full New Approach of Feature Extraction Method Based on the Raw Form and his Skeleton for Gujarati Handwritten Digits using Neural Networks Classifier
title_fullStr New Approach of Feature Extraction Method Based on the Raw Form and his Skeleton for Gujarati Handwritten Digits using Neural Networks Classifier
title_full_unstemmed New Approach of Feature Extraction Method Based on the Raw Form and his Skeleton for Gujarati Handwritten Digits using Neural Networks Classifier
title_sort new approach of feature extraction method based on the raw form and his skeleton for gujarati handwritten digits using neural networks classifier
publisher Universitatea Dunarea de Jos
series Analele Universităţii "Dunărea de Jos" Galaţi: Fascicula III, Electrotehnică, Electronică, Automatică, Informatică
issn 1221-454X
1221-454X
publishDate 2014-12-01
description This paper presents an optical character recognition (OCR) system for Gujarati handwritten digits. One may find so much of work for latin writing, arabic, chines, etc. but Gujarati is a language for which hardly any work is traceable especially for handwritten characters. Here in this work we have proposed a method of feature extraction based on the raw form of the character and his skeleton and we have shown the advantage of using this method over other approaches mentioned in this article.
topic Optical character recognition
neural network
feature extraction
Gujarati handwritten digits
skeletonization
classification
url http://www.ann.ugal.ro/eeai/archives/2014/1%20NEW%20APPROACH%20OF%20FEATURE%20EXTRACTION%20METHOD.pdf
work_keys_str_mv AT kmoro newapproachoffeatureextractionmethodbasedontherawformandhisskeletonforgujaratihandwrittendigitsusingneuralnetworksclassifier
AT mohammedfakir newapproachoffeatureextractionmethodbasedontherawformandhisskeletonforgujaratihandwrittendigitsusingneuralnetworksclassifier
AT belaidbouikhalene newapproachoffeatureextractionmethodbasedontherawformandhisskeletonforgujaratihandwrittendigitsusingneuralnetworksclassifier
AT rachidelyachi newapproachoffeatureextractionmethodbasedontherawformandhisskeletonforgujaratihandwrittendigitsusingneuralnetworksclassifier
AT baderdinneelkessab newapproachoffeatureextractionmethodbasedontherawformandhisskeletonforgujaratihandwrittendigitsusingneuralnetworksclassifier
_version_ 1725745783293083648