arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Computer Science · Human-Computer Interaction

Efficient CNN Based Sign Language Recognition System Using Optimization Technique

O. G. Manukumaar, Soumyalatha Naveen, U. M. Ashwinkumar

See on the map

Abstract

Convolutional Neural Network (CNN) - based sign language recognition has significant importance in bridging communication gaps between individuals who are deaf or hard of hearing and the general population. Sign language users can communicate effectively with each other through CNN-based sign language recognition systems that instantly translate sign language into written or spoken language. Integrating these technologies into public services, workplaces, educational institutions, and Internet platforms will improve everyone's participation and engagement without restrictions. People with hearing loss can live more freely using sign language recognition software. This article proposes developing a CNN architecture along with stochastic gradient descent optimizer in the framework which effectively identifies sign language. The experiment is conducted on various convolutional and pooling layer counts, filter sizes, and strides to find an architecture that balances accuracy and computational efficiency. For a dataset of hand gestures, it detected sign language with an accuracy of 98.5% using CNN. Such high accuracy indicates that the model is effective in learning and recognizing the patterns and features present in the dataset.

More like this