Real Time Braille & Sign Language Detection Using Artificial Neural Networks

Abstract

This research focuses on developing a real-time Braille and sign language detection system using artificial neural networks to enhance accessibility for visually and verbally impaired individuals. We developed novel datasets that can capture the real world context & employed advanced deep learning models, including Convolutional Neural Networks (CNNs) and YOLO, and utilized comprehensive data preprocessing techniques to ensure robustness. Our methodology achieved significant improvements in detection accuracy, validated through extensive performance metrics. The developed prototypes were tested in real-world scenarios, demonstrating practical effectiveness. This research advances assistive technologies, providing a foundation for future innovations and improving the quality of life for the targeted individuals.