Predictive Model to Detect Digital Eye Strain using Smart Goggle

Abstract

The proposed innovation is an effort to design a system which would be useful for anyone using a computer for prolonged hours such as students and IT professionals. Technology has made our lives easier by enabling us to work and access information from anywhere and anytime but it comes with its own set of limitations. One of the major limitations for computer users is the strain caused to eyes due to the persistent screen time. The proposed idea helps the users to identify the eye strain and fatigue caused by long working hours on the computer and enables them to take appropriate steps to prevent eye diseases caused by the same. Computer users experience various symptoms when their eyes get strained. Some of them can be easily noticed through video footage (traceable) but some of them cannot be trapped through visuals (non-traceable). We have proposed a model based on both traceable and non-traceable symptoms to predict eye strain or fatigue and send alerts to the users on the basis of the symptoms detected. The model has been developed by using deep learning algorithms. The model is trained by  feeding the data monitored through frames and input collected from users through live streaming.  The process starts with recording the user video using the webcam. The video is then used to extract frames and further factors/symptoms (traceable) are extracted from the frames. On the basis of extracted symptoms, further inputs are collected from the users (for non-traceable symptoms). The symptoms to be detected from the frames have been pre decided and the proposed algorithm detects whether the symptoms are present in the  recorded video.  In case symptoms are absent, the process is repeated for the time period for which the user is using the computer. In the presence of symptoms, the user is alerted about the eye strain and fatigue and guided to take further steps. This innovation can be developed into a useful system for a large population of computer users and prevent them from getting eye diseases as it would identify the cause of eye problems due to increased screen time in the current scenario.