Abstract
It is known that the blink duration is highly related to drowsiness, where the blink duration is the entire duration of one blink. Hence, it is important to measure the blink duration without any special wearable devices in various application, e.g., driver's monitoring. Although a Doppler sensor could be a key device to realize it, no such blink duration estimation method has been realized so far, since estimating the blink duration is difficult because of the low SNR (Signal to Noise Ratio) of the signal reflected from eyelids. In this paper, we propose a Doppler sensor-based method that estimates the duration, tblink, that is proportional to the actual blink duration. In the proposed method, a spectrogram is firstly calculated from a received signal, and then the timings when eyelids close and open are extracted from that. More specifically, tblink is calculated based on the center-of-gravity of the energy on a spectrogram. We conducted the experiments on five subjects to show the estimation accuracy of our proposed method. We confirmed that our method achieved the average correlation coefficient R of 0.95, furthermore, the average RMSE (Root Mean Square Error) of 46 ms.
| Original language | English |
|---|---|
| Title of host publication | 2018 IEEE International Conference on Communications, ICC 2018 - Proceedings |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Volume | 2018-May |
| ISBN (Print) | 9781538631805 |
| DOIs | |
| Publication status | Published - 2018 Jul 27 |
| Event | 2018 IEEE International Conference on Communications, ICC 2018 - Kansas City, United States Duration: 2018 May 20 → 2018 May 24 |
Other
| Other | 2018 IEEE International Conference on Communications, ICC 2018 |
|---|---|
| Country | United States |
| City | Kansas City |
| Period | 18/5/20 → 18/5/24 |
Fingerprint
ASJC Scopus subject areas
- Computer Networks and Communications
- Electrical and Electronic Engineering
Cite this
Doppler Sensor-Based Blink Duration Estimation by Spectrogram Analysis. / Yamamoto, Kohei; Toyoda, Kentaro; Ohtsuki, Tomoaki.
2018 IEEE International Conference on Communications, ICC 2018 - Proceedings. Vol. 2018-May Institute of Electrical and Electronics Engineers Inc., 2018. 8422999.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Doppler Sensor-Based Blink Duration Estimation by Spectrogram Analysis
AU - Yamamoto, Kohei
AU - Toyoda, Kentaro
AU - Ohtsuki, Tomoaki
PY - 2018/7/27
Y1 - 2018/7/27
N2 - It is known that the blink duration is highly related to drowsiness, where the blink duration is the entire duration of one blink. Hence, it is important to measure the blink duration without any special wearable devices in various application, e.g., driver's monitoring. Although a Doppler sensor could be a key device to realize it, no such blink duration estimation method has been realized so far, since estimating the blink duration is difficult because of the low SNR (Signal to Noise Ratio) of the signal reflected from eyelids. In this paper, we propose a Doppler sensor-based method that estimates the duration, tblink, that is proportional to the actual blink duration. In the proposed method, a spectrogram is firstly calculated from a received signal, and then the timings when eyelids close and open are extracted from that. More specifically, tblink is calculated based on the center-of-gravity of the energy on a spectrogram. We conducted the experiments on five subjects to show the estimation accuracy of our proposed method. We confirmed that our method achieved the average correlation coefficient R of 0.95, furthermore, the average RMSE (Root Mean Square Error) of 46 ms.
AB - It is known that the blink duration is highly related to drowsiness, where the blink duration is the entire duration of one blink. Hence, it is important to measure the blink duration without any special wearable devices in various application, e.g., driver's monitoring. Although a Doppler sensor could be a key device to realize it, no such blink duration estimation method has been realized so far, since estimating the blink duration is difficult because of the low SNR (Signal to Noise Ratio) of the signal reflected from eyelids. In this paper, we propose a Doppler sensor-based method that estimates the duration, tblink, that is proportional to the actual blink duration. In the proposed method, a spectrogram is firstly calculated from a received signal, and then the timings when eyelids close and open are extracted from that. More specifically, tblink is calculated based on the center-of-gravity of the energy on a spectrogram. We conducted the experiments on five subjects to show the estimation accuracy of our proposed method. We confirmed that our method achieved the average correlation coefficient R of 0.95, furthermore, the average RMSE (Root Mean Square Error) of 46 ms.
UR - http://www.scopus.com/inward/record.url?scp=85051439925&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85051439925&partnerID=8YFLogxK
U2 - 10.1109/ICC.2018.8422999
DO - 10.1109/ICC.2018.8422999
M3 - Conference contribution
AN - SCOPUS:85051439925
SN - 9781538631805
VL - 2018-May
BT - 2018 IEEE International Conference on Communications, ICC 2018 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
ER -