Efficient Sign Language Classification Utilizing Neural Networks

View at publisher Summarise with VIRAT
Authors: Basel A. Dabwan, Majda Omer Elbasheer, Arwa M. Eldhai, Nabila A. Saeed, Soad M. Almula, Yahya A. Ali
Cited by: 0
References: 14

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The Translation System (TS) is a crucial innovation aimed at supporting people with disabilities by enhancing their ability to communicate effectively within society. As computers become increasingly embedded in everyday activities, progress in human-computer interaction has brought significant advantages to individuals who are mute or deaf. Gesture language serves as a fundamental medium for closing communication gaps faced by those with hearing impairments. Our proposed system presents an intelligent translation model capable of converting sign language gestures into written text, thereby enabling smooth interaction between individuals with speech or hearing difficulties and the wider community. The system is built on a RegNet-inspired neural architecture and trained using a comprehensive Sign Language Dataset A total of 27,455 instances are used for training, while 7,172 are allocated for testing. With an outstanding accuracy rate of 99.86%, as illustrated in the accuracy, confusion matrix, and class distribution figures, the model proves to be highly efficient and dependable in translating sign language gestures into text form.

Fields of Study & Subjects

Hand Gesture Recognition Systems Hearing Impairment and Communication Human Pose and Action Recognition

Publication details

Journal / source
International Conference on Communication, Computing and Electronics Systems (ICCCES)
Pages
1877–1881
Publication date
21 January 2026
Language
EN