Title :
Blink detection using Doppler sensor
Author :
Tamba, Chihiro ; Tomii, Shoichiro ; Ohtsuki, Tomoaki
Author_Institution :
Grad. Sch. of Sci. & Technol., Keio Univ., Yokohama, Japan
Abstract :
There have been a lot of attention paid to health and safety, such as dry eye syndrome or falling asleep while driving. Blinks are the physiological signal that reflects the degree of concentration or drowsiness, and thus the blink detection system is expected to get blink information. In many blink detection systems it is necessary to contact the equipment or use with a camera and the feeling of oppression and resistance will arise. In this paper, we propose the contactless blink detection system with Doppler sensor. With the setting of the threshold for each person and the classification of a blink and non-blink, the proposed system can detect and count blinks. It is shown that the system achieves the recall rate of 92 % as for the detection accuracy by experiment.
Keywords :
image sensors; object detection; Doppler sensor; blink detection; blink detection system; contactless blink detection system; dry eye syndrome; physiological signal; visual display terminal work; Accuracy; Cameras; Doppler effect; Fatigue; Feature extraction; Noise reduction; Smoothing methods;
Conference_Titel :
Personal, Indoor, and Mobile Radio Communication (PIMRC), 2014 IEEE 25th Annual International Symposium on
DOI :
10.1109/PIMRC.2014.7136522