CNN Algorithm with SIFT to Enhance the Arabic Sign Language Recognition
Manar Hamza Bashaa1, Faezah Hamad Almasoudy2, Noor S. Sagheer3, Wasan Mueti Hadi4
1Manar Hamza Bashaa, Department of Computer Science, College of Computer Science and Information Technology, Kerbala University, Kerbala, Iraq.
2Faezah Hamad Almasoudy, Department of Animals Production, College of Agriculture, Kerbala University, Kerbala, Iraq.
3Noor S. Sagheer, Department of Computer Science, College of Computer Science and Information Technology, Kerbala University, Kerbala, Iraq.
4Wasan Mueti Hadi, Department of Computer Science, College of Computer Science and Information Technology, Kerbala University, Kerbala, Iraq.
Manuscript received on 24 August 2024 | Revised Manuscript received on 02 September 2024 | Manuscript Accepted on 15 September 2024 | Manuscript published on 30 September 2024 | PP: 12-17 | Volume-12 Issue-10, September 2024 | Retrieval Number: 100.1/ijese.I258412100924 | DOI: 10.35940/ijese.I2584.12100924
Open Access | Editorial and Publishing Policies | Cite | Zenodo | OJS | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Sign language is used as a primary means of communication by millions of people who suffer from hearing problems. The unhearing people used visual language to interact with each other, Represented in sign language. There are features that the hearing-impaired use to communicate with each other, which are often difficult for people with normal hearing to understand. Therefore, deaf people will struggle to interact with society. This research aims to introduce a system for recognising hand gestures in Arabic Sign Language (ArSL) by training a Convolutional Neural Network (CNN) on images of ArSL gestures provided by the University of Prince Mohammad Bin Fahd, Saudi Arabia. A Scale Invariant Feature Transform (SIFT) algorithm is used to create feature vectors that contain shape, finger position, size, centre points of the palm, and hand margin by extracting important features from images of ArSL and transforming them into vector points. The accuracy of the proposed system is 97% using SIFT with CNN, and is nearly equal to 94.8% without SIFT. Finally, the proposed system was tested on a group of individuals, and its effectiveness was demonstrated through their observations.
Keywords: Arabic Language, CNN Classification, Deep Learning, Image Classification, SIFT.
Scope of the Article: Artificial Intelligence & Methods