Related Experiment Video
Updated: Aug 13, 2026
DeepOmicsAE: Representing Signaling Modules in Alzheimer's Disease with Deep Learning Analysis of Proteomics, Metabolomics, and Clinical Data
Published on: December 15, 2023
Dimensionality Reduction for Offline Alphabet Arabic Sign Language Recognition using Deep Learning
Sari Awwad1, Subhieh M El-Salhi2, Bashar Igried1
1Department of Computer Science and Applications, Faculty of Prince Al-Hussein Bin Abdullah II for Information Technology, The Hashemite University, Zarqa, Jordan.
This study developed an offline Arabic Sign Language (ArSL) recognition system using Principal Component Analysis (PCA), Scale-Invariant Feature Transform (SIFT), and Convolutional Neural Networks (CNNs). The system achieved 86.64% accuracy, improving communication for hearing-impaired individuals.
Area of Science:
- Computer Vision
- Artificial Intelligence
- Human-Computer Interaction
Background:
- Arabic Sign Language (ArSL) recognition technology lags behind other sign languages like American Sign Language (ASL).
- This technological gap limits communication accessibility for the deaf community in Arabic-speaking regions, especially in offline settings with limited computing power.
Purpose of the Study:
- To develop a robust, offline recognition system for ArSL.
- To enhance communication accessibility for individuals with hearing impairments in Arabic-speaking regions.
Main Methods:
- Utilized Principal Component Analysis (PCA) for dimensionality reduction and Scale-Invariant Feature Transform (SIFT) for feature extraction.
- Employed Convolutional Neural Networks (CNNs) for gesture classification on a curated ArSL dataset.
- Implemented preprocessing techniques including normalization, contrast enhancement, and noise reduction.
Main Results:
- Achieved a recognition accuracy of 86.64%, outperforming traditional SIFT+SVM models (84.45%).
- Demonstrated that integrating PCA and SIFT improved recognition efficiency and reduced model complexity.
- Highlighted the superior adaptability and precision of deep learning methods (CNNs) for ArSL gesture recognition.
Conclusions:
- Presents a functional offline ArSL recognition system designed to improve communication, education, and social inclusion.
- The developed system offers a viable solution for enhancing the quality of life for individuals with hearing impairments in Arabic-speaking communities.
- This research contributes to bridging the technological gap in sign language recognition for underrepresented linguistic communities.