Novel Feature Extraction and Classification Technique for Sensor-Based Continuous Arabic Sign Language Recognition

This paper proposes a novel approach to continuous Arabic Sign Language recognition. We use a dataset which contains 40 sentences composed from 80 sign language words. The dataset is collected using sensor-based gloves. We propose a novel set of features suitable for sensor readings based on covaria...

Full description

Saved in:
Bibliographic Details
Main Author: Tuffaha, Mohammed (author)
Other Authors: Shanableh, Tamer (author), Assaleh, Khaled (author)
Format: article
Published: 2015
Subjects:
Online Access:http://hdl.handle.net/11073/21379
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1864513434497843200
author Tuffaha, Mohammed
author2 Shanableh, Tamer
Assaleh, Khaled
author2_role author
author
author_facet Tuffaha, Mohammed
Shanableh, Tamer
Assaleh, Khaled
author_role author
dc.creator.none.fl_str_mv Tuffaha, Mohammed
Shanableh, Tamer
Assaleh, Khaled
dc.date.none.fl_str_mv 2015
2021-03-18T08:56:09Z
2021-03-18T08:56:09Z
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv Tuffaha M., Shanableh T., Assaleh K. (2015) Novel Feature Extraction and Classification Technique for Sensor-Based Continuous Arabic Sign Language Recognition. In: Arik S., Huang T., Lai W., Liu Q. (eds) Neural Information Processing. ICONIP 2015. Lecture Notes in Computer Science, vol 9492. Springer, Cham. https://doi.org/10.1007/978-3-319-26561-2_35
978-3-319-26561-2
http://hdl.handle.net/11073/21379
10.1007/978-3-319-26561-2_35
dc.language.none.fl_str_mv en_US
dc.publisher.none.fl_str_mv Springer
dc.relation.none.fl_str_mv https://doi.org/10.1007/978-3-319-26561-2_35
dc.subject.none.fl_str_mv Sign language recognition
Feature extraction
Sensor-based gloves
Pattern classification
dc.title.none.fl_str_mv Novel Feature Extraction and Classification Technique for Sensor-Based Continuous Arabic Sign Language Recognition
dc.type.none.fl_str_mv Peer-Reviewed
Preprint
info:eu-repo/semantics/publishedVersion
info:eu-repo/semantics/article
description This paper proposes a novel approach to continuous Arabic Sign Language recognition. We use a dataset which contains 40 sentences composed from 80 sign language words. The dataset is collected using sensor-based gloves. We propose a novel set of features suitable for sensor readings based on covariance, smoothness, entropy and uniformity. We also propose a novel classification approach based on a modified polynomial classifier suitable for sequential data. The proposed classification scheme is modified to take into account the context of the feature vectors prior to classification. This is achieved through the filtering of predicted class labels using median and mode filtering. The proposed work is compared against a vision-based solution. The proposed solution is found to outperform the vision-based solution as it yields an improved sentence recognition rate of 85 %.
format article
id aus_fe0b686de1ca018cd8e9df7853a3e0e7
identifier_str_mv Tuffaha M., Shanableh T., Assaleh K. (2015) Novel Feature Extraction and Classification Technique for Sensor-Based Continuous Arabic Sign Language Recognition. In: Arik S., Huang T., Lai W., Liu Q. (eds) Neural Information Processing. ICONIP 2015. Lecture Notes in Computer Science, vol 9492. Springer, Cham. https://doi.org/10.1007/978-3-319-26561-2_35
978-3-319-26561-2
10.1007/978-3-319-26561-2_35
language_invalid_str_mv en_US
network_acronym_str aus
network_name_str aus
oai_identifier_str oai:repository.aus.edu:11073/21379
publishDate 2015
publisher.none.fl_str_mv Springer
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
spelling Novel Feature Extraction and Classification Technique for Sensor-Based Continuous Arabic Sign Language RecognitionTuffaha, MohammedShanableh, TamerAssaleh, KhaledSign language recognitionFeature extractionSensor-based glovesPattern classificationThis paper proposes a novel approach to continuous Arabic Sign Language recognition. We use a dataset which contains 40 sentences composed from 80 sign language words. The dataset is collected using sensor-based gloves. We propose a novel set of features suitable for sensor readings based on covariance, smoothness, entropy and uniformity. We also propose a novel classification approach based on a modified polynomial classifier suitable for sequential data. The proposed classification scheme is modified to take into account the context of the feature vectors prior to classification. This is achieved through the filtering of predicted class labels using median and mode filtering. The proposed work is compared against a vision-based solution. The proposed solution is found to outperform the vision-based solution as it yields an improved sentence recognition rate of 85 %.Springer2021-03-18T08:56:09Z2021-03-18T08:56:09Z2015Peer-ReviewedPreprintinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfTuffaha M., Shanableh T., Assaleh K. (2015) Novel Feature Extraction and Classification Technique for Sensor-Based Continuous Arabic Sign Language Recognition. In: Arik S., Huang T., Lai W., Liu Q. (eds) Neural Information Processing. ICONIP 2015. Lecture Notes in Computer Science, vol 9492. Springer, Cham. https://doi.org/10.1007/978-3-319-26561-2_35978-3-319-26561-2http://hdl.handle.net/11073/2137910.1007/978-3-319-26561-2_35en_UShttps://doi.org/10.1007/978-3-319-26561-2_35oai:repository.aus.edu:11073/213792024-08-22T12:08:42Z
spellingShingle Novel Feature Extraction and Classification Technique for Sensor-Based Continuous Arabic Sign Language Recognition
Tuffaha, Mohammed
Sign language recognition
Feature extraction
Sensor-based gloves
Pattern classification
status_str publishedVersion
title Novel Feature Extraction and Classification Technique for Sensor-Based Continuous Arabic Sign Language Recognition
title_full Novel Feature Extraction and Classification Technique for Sensor-Based Continuous Arabic Sign Language Recognition
title_fullStr Novel Feature Extraction and Classification Technique for Sensor-Based Continuous Arabic Sign Language Recognition
title_full_unstemmed Novel Feature Extraction and Classification Technique for Sensor-Based Continuous Arabic Sign Language Recognition
title_short Novel Feature Extraction and Classification Technique for Sensor-Based Continuous Arabic Sign Language Recognition
title_sort Novel Feature Extraction and Classification Technique for Sensor-Based Continuous Arabic Sign Language Recognition
topic Sign language recognition
Feature extraction
Sensor-based gloves
Pattern classification
url http://hdl.handle.net/11073/21379