| A. Kanavos, O. Papadimitriou, Ph. Mylonas, M. Maragoudakis |
| Enhancing Sign Language Recognition using Deep Convolutional Neural Networks |
| 14th International Conference on Information, Intelligence, Systems and Applications (IISA 2023), 10-12 July 2023, Volos, Greece |
|
ABSTRACT
|
| The progress in sensing technologies and AI algorithms has opened up numerous possibilities for the development of diverse applications that aim to fulfill the requirements of people who are deaf or hard of hearing. The significance of sign language cannot be overstated for individuals who suffer from hearing and speaking disabilities. The purpose of this research is to explore digital image processing and machine learning methods that can be utilized to develop a sign language dataset in a productive manner and also to create a sign language interface system that uses a Convolutional Neural Network to decipher hand gestures and poses and convert them into natural language. The neural network developed for this study is a Convolutional Neural Network (CNN) that improves the accuracy of predicting the American Sign Language alphabet. Despite the variations in the conditions and size of the dataset, it was able to attain an exceptional accuracy rate of 98.73%, while also demonstrating a low loss value of 0.0539.
|
| 10 July , 2023 |
| A. Kanavos, O. Papadimitriou, Ph. Mylonas, M. Maragoudakis, "Enhancing Sign Language Recognition using Deep Convolutional Neural Networks", 14th International Conference on Information, Intelligence, Systems and Applications (IISA 2023), 10-12 July 2023, Volos, Greece |
[  PDF] [
BibTex] [
Print] [
 Back] |
