Recognition of Off-line printed Arabic text Using Hidden Markov Models

This paper describes a technique for automatic recognition of off-line printed Arabic text using Hidden Markov Models. In this work different sizes of overlapping and non-overlapping hierarchical windows to generate 16 features from each vertical sliding strip are used. We experimented with all test...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Al-Muhtaseb, Husni A. (author)
مؤلفون آخرون: Mahmoud, Sabri A. (author), Qahwaji, Rami S. (author), unknown (author)
التنسيق: article
منشور في: 2008
الموضوعات:
الوصول للمادة أونلاين:https://eprints.kfupm.edu.sa/id/eprint/17670/1/Recognition_of_Off-line_printed_Arabic_text_Using_Hidden_Markov_Models_Abstract.pdf
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513394638323712
author Al-Muhtaseb, Husni A.
author2 Mahmoud, Sabri A.
Qahwaji, Rami S.
unknown
author2_role author
author
author
author_facet Al-Muhtaseb, Husni A.
Mahmoud, Sabri A.
Qahwaji, Rami S.
unknown
author_role author
dc.creator.none.fl_str_mv Al-Muhtaseb, Husni A.
Mahmoud, Sabri A.
Qahwaji, Rami S.
unknown
dc.date.none.fl_str_mv 2008-12
2020
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv https://eprints.kfupm.edu.sa/id/eprint/17670/1/Recognition_of_Off-line_printed_Arabic_text_Using_Hidden_Markov_Models_Abstract.pdf
(2008) Recognition of Off-line printed Arabic text Using Hidden Markov Models. SIGNAL PROCESSING, 88 (12). pp. 2902-2912.
dc.language.none.fl_str_mv en
dc.publisher.none.fl_str_mv ELSEVIER SCIENCE BV
dc.relation.none.fl_str_mv https://eprints.kfupm.edu.sa/id/eprint/17670/
dc.rights.*.fl_str_mv info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Computer
Electrical
dc.title.none.fl_str_mv Recognition of Off-line printed Arabic text Using Hidden Markov Models
dc.type.none.fl_str_mv Article
PeerReviewed
info:eu-repo/semantics/publishedVersion
info:eu-repo/semantics/article
description This paper describes a technique for automatic recognition of off-line printed Arabic text using Hidden Markov Models. In this work different sizes of overlapping and non-overlapping hierarchical windows to generate 16 features from each vertical sliding strip are used. We experimented with all tested fonts (viz. Arial, Tahoma, Akhbar, Thuluth, Naskh, Simplified Arabic, Andalus, and Traditional Arabic). It was experimentally proven that different fonts have their highest recognition rates at different numbers of states (5 or 7) and codebook sizes (128 or 256). Arabic text is cursive, and each character may have up to 4 different shapes based in its location in a word. We decided to consider each shape as a different class hence resulting in a total of 126 classes. The achieved average recognition rates (using 126 classes and 16 features for each vertical strip of three pixels width) were between 98.08% for Thuluth and 99.89% for Arial. The main contributions of this work are the novel hierarchical sliding window technique, and using 16 features only for each sliding window. Each shape of the Arabic characters is considered as a separate class, bypassing the need for segmenting Arabic text, and is applicable to other languages.
eu_rights_str_mv openAccess
format article
id KFUPM_fd42e18b498f9d659439586b8820e6df
identifier_str_mv (2008) Recognition of Off-line printed Arabic text Using Hidden Markov Models. SIGNAL PROCESSING, 88 (12). pp. 2902-2912.
language_invalid_str_mv en
network_acronym_str KFUPM
network_name_str King Fahd University of Petroleum and Minerals
oai_identifier_str oai::17670
publishDate 2008
publisher.none.fl_str_mv ELSEVIER SCIENCE BV
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
spelling Recognition of Off-line printed Arabic text Using Hidden Markov ModelsAl-Muhtaseb, Husni A.Mahmoud, Sabri A.Qahwaji, Rami S.unknownComputerElectricalThis paper describes a technique for automatic recognition of off-line printed Arabic text using Hidden Markov Models. In this work different sizes of overlapping and non-overlapping hierarchical windows to generate 16 features from each vertical sliding strip are used. We experimented with all tested fonts (viz. Arial, Tahoma, Akhbar, Thuluth, Naskh, Simplified Arabic, Andalus, and Traditional Arabic). It was experimentally proven that different fonts have their highest recognition rates at different numbers of states (5 or 7) and codebook sizes (128 or 256). Arabic text is cursive, and each character may have up to 4 different shapes based in its location in a word. We decided to consider each shape as a different class hence resulting in a total of 126 classes. The achieved average recognition rates (using 126 classes and 16 features for each vertical strip of three pixels width) were between 98.08% for Thuluth and 99.89% for Arial. The main contributions of this work are the novel hierarchical sliding window technique, and using 16 features only for each sliding window. Each shape of the Arabic characters is considered as a separate class, bypassing the need for segmenting Arabic text, and is applicable to other languages.ELSEVIER SCIENCE BV2008-122020ArticlePeerReviewedinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfhttps://eprints.kfupm.edu.sa/id/eprint/17670/1/Recognition_of_Off-line_printed_Arabic_text_Using_Hidden_Markov_Models_Abstract.pdf (2008) Recognition of Off-line printed Arabic text Using Hidden Markov Models. SIGNAL PROCESSING, 88 (12). pp. 2902-2912. enhttps://eprints.kfupm.edu.sa/id/eprint/17670/info:eu-repo/semantics/openAccessoai::176702019-11-01T14:09:30Z
spellingShingle Recognition of Off-line printed Arabic text Using Hidden Markov Models
Al-Muhtaseb, Husni A.
Computer
Electrical
status_str publishedVersion
title Recognition of Off-line printed Arabic text Using Hidden Markov Models
title_full Recognition of Off-line printed Arabic text Using Hidden Markov Models
title_fullStr Recognition of Off-line printed Arabic text Using Hidden Markov Models
title_full_unstemmed Recognition of Off-line printed Arabic text Using Hidden Markov Models
title_short Recognition of Off-line printed Arabic text Using Hidden Markov Models
title_sort Recognition of Off-line printed Arabic text Using Hidden Markov Models
topic Computer
Electrical
url https://eprints.kfupm.edu.sa/id/eprint/17670/1/Recognition_of_Off-line_printed_Arabic_text_Using_Hidden_Markov_Models_Abstract.pdf