Identifying and preventing computer vision syndrome (CVS)

Abstract

The rapid expansion of digital device usage has led to a surge in Computer Vision Syndrome (CVS), significantly affecting users' eye health, productivity, and overall well-being. Existing CVS detection methods rely primarily on subjective self-assessments,which lack real-time monitoring and objective symptom evaluation. This research proposes a real-time CVS detection system utilizing Convolutional Neural Networks (CNNs) and computer vision techniques to analyse critical indicators such as blink rate, facial expressions, and head posture. By leveraging deep learning, the system provides timely alerts, encouraging preventive measures to reduce eye strain and associated discomfort. Experimental evaluations demonstrate that the proposed CNN-based model achieves 94% accuracy, effectively identifying CVS symptoms while maintaining real-time performance. The system is designed to adapt to varying environmental conditions, ensuring scalability, usability, and seamless integration into daily digital interactions. By introducing an objective, and real-time monitoring solution, this study contributes to advancing health-focused computer vision applications, offering a practical, data-driven approach to mitigating the impact of prolonged screen exposure.