|
Abstract
|
The proposed method uses computer vision and deep learning techniques to provide a real-time diagnosis model, in contrast to the most driver drowsiness detection methods that are based on biological approach (for example, electroencephalogram and electrooculogram), which require special devices and are expensive. It uses deep learning networks to classify data and handles automatic face assessment, eye aspect ratio assessment, and alertness monitoring for users. Particularly intriguing characteristics of a human position can be found in measurements like gaze, head movement, and eye blink rate [6]. Additionally, blink detection is crucial for systems that track human operator alertness, warn against dry eyes and computer vision syndromes, facilitate communication for people with disabilities, or provide anti-spoofing security for facial recognition systems [7]. However, very little research has been done to apply blink analysis to an operator's safety-critical jobs in process plants, such as those in the nuclear and oil and gas industries, where driving while fatigued is a serious offense. Therefore, the purpose of this research is to create an effective model based on eye recognition and blink analysis for operator sleepiness categorization in industrial plants in real time. You can tell if someone is awake or asleep based on how their eyes are functioning. Additionally, it is anticipated that it will contribute to the field of operational performance state (OFS), which is described as the variable capacity of the operator to perform work effectively in response to environmental tasks and demands, and under typical conditions.
|