Encouraged by Her Excellency the teaching staff at our university, the President of our University, Prof. Dr. Alia Abbas Ali Al-Attar, received in her office Mr. Ahmed Saeed Ibrahim Al-Obaidi, who is one of our university’s teachers in the Studies and Planning Department, and during the meeting, he presented to Her Excellency a copy of his published book, which is a joint book with the researchers:
Prof. Dr. Raed Raed Omar Al-Nama, a lecturer in the Graduate Studies Department at the University Presidency
D. Tink Han (Birkbeck, University of London / UK)
The title of the book:
Interpreting the Sign Language of the Arabic Alphabet””
book’s content:
This book focused on Sign Language (SL), which is an important means of communication between people. It is a primary language for people with disabilities who are deaf and dumb, and can be considered their mother tongue. Hand gestures make up the non-verbal communication of this language.
Emphasis was placed on translating Arabic Alphabet Images (ASAI) in this book, and as a study environment, the Iraqi Sign Language (IrSL) alphabet was implemented with the help of specialists at the “Al-Amal Institute for the Deaf and Dumb”. A new ASAI dataset was collected from different images of hand gestures. In addition, a deep learning (DL) model named Deep Arabic Alphabet (DASA) has been proposed. It is an improved version of Convolutional Neural Network (CNN) and has the ability to interpret ASAI efficiently.
Also in this book a new approach to Interpretation of Arabic Alphabet Signs (ASAS) is proposed, in which IrSL is considered as a study environment as well. A new ASAS dataset was obtained from different hand gesture cues. The dataset was created using a glove paired with flexion sensors to sense the movement of each finger; Each sensor gives a variable value based on the percentage of curvature of the finger. The sensors connect to an Arduino controller which is then connected to a computer to transfer data. Moreover a DL method called Deep Recurrent Alphabetic Sign Language (DRASL) which uses several Deep Recurrent Neural Networks (DRNN) has been adopted. Also, an educational application for the deaf and dumb was created by taking advantage of ASAS and DRASL to translate and show the Arabic alphabet.
Publication and editions of the book:
This book has been published by LAP Lambert Academic Publishing and is in English. Other sources for this book have also been documented for different languages: Russian, Spanish, German, Italian, French and Portuguese.