Journal
2021 IEEE INTERNATIONAL CONFERENCE ON CONSUMER ELECTRONICS AND COMPUTER ENGINEERING (ICCECE)
Volume -, Issue -, Pages 670-675Publisher
IEEE
DOI: 10.1109/ICCECE51280.2021.9342080
Keywords
fatigue driving; millimeter wave radar; respiration; heartbeat logistics regression
Categories
Funding
- National Natural Science Foundation of China [61301012, 61471140]
- Sci-tech Innovation Foundation of Harbin [2016RALGJ001]
Ask authors/readers for more resources
This paper proposes a method for fatigue driving detection based on FMCW radar, achieving an accuracy of up to 85% through the fusion of multiple physiological indicators. It provides a new possibility for non-contact detection and multi-index measurement.
Fatigue driving often causes serious traffic accidents and heavy casualties. In order to detect fatigue driving, the application based on driver's physiological characteristics have been presented and investigated. However, the existing methods are usually contacting and the detection environment is singular. The accuracy under face occlusion is abs s loss and some fatigue recognition indicators are missing. Therefore, this paper propos-method of fatigue driving detection based on Frequency Modulated continuous Wase (FMCW) radar. An millimeter ave(mmWave) AWRI 642 radar sensor was chosen, and the platform of fatigue driving detection was designed and built. Respiration and heartbeat signals were acquired, separated and preprocessed_ In addition, logistics regression w as utilized in the fatigue driving judgment algorithm Multiple indicators of heart rate frequency, respiration frequency, hear rate amplitude and respiration amplitude were fused. The results showed that the accuracy of determining driving fatigue was up to 85 % Our proposed method realizes the accurate non-contacting detection of respiration and heartbeat signals of human beings. It provides all-weather and multi-index measurement, is hich is suitable to must driving environments.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available