Optimizing Web Access for Hand Disability Sufferers Through Blink Technology and Deep Learning with CNN/Tensor Flow

Abstract

This study aims to enhance web accessibility for individuals with hand disabilities by leveraging eye-blinking technology and a deep learning model based on Convolutional Neural Networks (CNN). The primary focus is on developing a system that enables interaction with web interfaces through eye blinks. The CNN model is used to detect key elements on web pages, such as buttons and links, which can then be accessed via eye blink input. The dataset includes images of web interfaces and eye blink data used to train and test the model. Results demonstrate that the system significantly improves web accessibility with high detection accuracy and responsive interaction. User evaluations indicate that the system effectively facilitates access for those with hand limitations, offering a valuable alternative to enhance their web experience. This research contributes to the development of more inclusive digital accessibility solutions and has the potential to improve the quality of life for individuals with hand disabilities.