FatigueWatcher: Interactive Fatigue Detection for Personal Computer and Mobile Device
We introduce our FatigurWatcher that detects fatigue of PC and mobile device users. To achieve our system, we use a web camera embedded into most of PC and mobile devices. The system captures user’s face in front of the devices and then quantifies face characters such as blinks, yawns, and facial inclinations every second. Although our fatigue detection is simple, we could achieve enough accuracy to use it for a wide variety of applications. Moreover, through our initial evaluations, we were able to determine that the number of blinks is the strongest indicator to detect possible fatigue. By running Bayesian estimations from the obtained data, we were able to determine that this system can detect fatigue with an accuracy of 87.5%. In this paper, we describe our FatigureWatch focusing on the concept, implementation, and evaluation.
KeywordsFatigue detection Facial features Image processing Health care Personal computer Mobile device Interactive system Bayesian estimation
- 1.Bao, D., Yang, Z., Song, Y.: Projection function for driver fatigue monitoring with monocular camera. In: Proceedings of ACM SAC, pp. 11–15, March 2007. https://doi.org/10.1145/1244002.1244022
- 2.Alhazmi, S., Saini, M., El Saddik, M.: Multimedia fatigue detection for adaptive infotainment. In: Proceedings of ACM HCMC, pp. 15–24, October 2015. https://doi.org/10.1145/2810397.2810440
- 3.King, D.E.: Dlib-ml: a machine learning toolkit. J. Mach. Learn. Res. 10, 1755–1758 (2009)Google Scholar