Efficient deep learning models based on tension techniques for sign language recognition
Communication by speaking prevails among the various ways of self-expression and communication between people. Speech presents a significant challenge for some disabled people, such as deaf people, deaf and hard of hearing, dumb and wordless persons. Therefore, these people rely on sign language to...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-11-01
|
| Series: | Intelligent Systems with Applications |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2667305323001096 |
| _version_ | 1797526182238879744 |
|---|---|
| author | Nehal F. Attia Mohamed T. Faheem Said Ahmed Mahmoud A.M. Alshewimy |
| author_facet | Nehal F. Attia Mohamed T. Faheem Said Ahmed Mahmoud A.M. Alshewimy |
| author_sort | Nehal F. Attia |
| collection | DOAJ |
| description | Communication by speaking prevails among the various ways of self-expression and communication between people. Speech presents a significant challenge for some disabled people, such as deaf people, deaf and hard of hearing, dumb and wordless persons. Therefore, these people rely on sign language to interact with others. Sign language is a system of movements and visual messages that ensure the integration of these individuals into groups that communicate vocally. On the other side, it is necessary to understand these individuals' gestures and linguistic semantics. The main objective of this work is to establish a new model that enhances the performance of the existing paradigms used for sign language recognition. This study developed three improved deep-learning models based on YOLOv5x and attention methods for recognizing the alphabetic and numeric information hand gestures convey. These models were evaluated using the MU HandImages ASL and OkkhorNama: BdSL datasets. The proposed models exceed those found in the literature, where the accuracy reached 98.9 % and 97.6 % with the MU HandImages ASL dataset and the OkkhorNama: BdSL dataset, respectively. The proposed models are light and fast enough to be used in real-time ASL recognition and to be deployed on any edge-based platform. |
| first_indexed | 2024-03-10T09:25:36Z |
| format | Article |
| id | doaj.art-5cf4b9e222d442d19647c746304dfd05 |
| institution | Directory Open Access Journal |
| issn | 2667-3053 |
| language | English |
| last_indexed | 2024-03-10T09:25:36Z |
| publishDate | 2023-11-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Intelligent Systems with Applications |
| spelling | doaj.art-5cf4b9e222d442d19647c746304dfd052023-11-22T04:49:37ZengElsevierIntelligent Systems with Applications2667-30532023-11-0120200284Efficient deep learning models based on tension techniques for sign language recognitionNehal F. Attia0Mohamed T. Faheem Said Ahmed1Mahmoud A.M. Alshewimy2Computer and Automatic Control Department, Faculty of Engineering, Tanta University, Tanta, Egypt; Computer Engineering Department, Faculty of Engineering, Pharos University, Alexandria, Egypt; Corresponding author at: Computer and Automatic Control Department, Faculty of Engineering, Tanta University, Tanta, Egypt.Computer and Automatic Control Department, Faculty of Engineering, Tanta University, Tanta, EgyptComputer and Automatic Control Department, Faculty of Engineering, Tanta University, Tanta, EgyptCommunication by speaking prevails among the various ways of self-expression and communication between people. Speech presents a significant challenge for some disabled people, such as deaf people, deaf and hard of hearing, dumb and wordless persons. Therefore, these people rely on sign language to interact with others. Sign language is a system of movements and visual messages that ensure the integration of these individuals into groups that communicate vocally. On the other side, it is necessary to understand these individuals' gestures and linguistic semantics. The main objective of this work is to establish a new model that enhances the performance of the existing paradigms used for sign language recognition. This study developed three improved deep-learning models based on YOLOv5x and attention methods for recognizing the alphabetic and numeric information hand gestures convey. These models were evaluated using the MU HandImages ASL and OkkhorNama: BdSL datasets. The proposed models exceed those found in the literature, where the accuracy reached 98.9 % and 97.6 % with the MU HandImages ASL dataset and the OkkhorNama: BdSL dataset, respectively. The proposed models are light and fast enough to be used in real-time ASL recognition and to be deployed on any edge-based platform.http://www.sciencedirect.com/science/article/pii/S2667305323001096American sign language (ASL)YOLOv5Object recognitionComputer visionConvolutional block attention module (CBAM)Squeeze-and-excitation (SE) |
| spellingShingle | Nehal F. Attia Mohamed T. Faheem Said Ahmed Mahmoud A.M. Alshewimy Efficient deep learning models based on tension techniques for sign language recognition Intelligent Systems with Applications American sign language (ASL) YOLOv5 Object recognition Computer vision Convolutional block attention module (CBAM) Squeeze-and-excitation (SE) |
| title | Efficient deep learning models based on tension techniques for sign language recognition |
| title_full | Efficient deep learning models based on tension techniques for sign language recognition |
| title_fullStr | Efficient deep learning models based on tension techniques for sign language recognition |
| title_full_unstemmed | Efficient deep learning models based on tension techniques for sign language recognition |
| title_short | Efficient deep learning models based on tension techniques for sign language recognition |
| title_sort | efficient deep learning models based on tension techniques for sign language recognition |
| topic | American sign language (ASL) YOLOv5 Object recognition Computer vision Convolutional block attention module (CBAM) Squeeze-and-excitation (SE) |
| url | http://www.sciencedirect.com/science/article/pii/S2667305323001096 |
| work_keys_str_mv | AT nehalfattia efficientdeeplearningmodelsbasedontensiontechniquesforsignlanguagerecognition AT mohamedtfaheemsaidahmed efficientdeeplearningmodelsbasedontensiontechniquesforsignlanguagerecognition AT mahmoudamalshewimy efficientdeeplearningmodelsbasedontensiontechniquesforsignlanguagerecognition |