Design of a Fatigue Detection System Based on Indirect-Contact ECG Measurement

• Main Authors: Cong-Yu - Wang, 王聰瑜
• Other Authors: Yuan-Hsiang Lin
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/01086909194384936466

碩士 === 國立臺灣科技大學 === 電子工程系 === 105 === As number of news have reported traffic accidents caused by driving fatigue, and some research study has also pointed out that the driving accidents due to driving fatigue, however, is as malignant as drink driving. When driver is affected by lack of sleep and a long-term driving, the driver tends to suffer from fatigue, poor judgment and slow reaction time. And a critical car accident may have occurred if the driver doesn’t take a rest at the time. Therefore, the aim of the thesis is to develop a fatigue detection system based on ECG monitoring. To observe the variation of the vital signs when driving, in order to provide alerts and related information to the drivers by portable devices in real-time when they are exhausted or abnormal. In this research, an indirect-contact ECG measurement is used to detect R-peak positions and the heart rate variability (HRV, which also refers to R-R Interval array). Then analyze the data in time-domain and frequency-domain based on standards of measurement and physiological interpretation of HRV. In frequency domain, an internal digital signal processor of the microcontroller is used to access the spectral information of the HRV by the Welch’s method. Furthermore, we can apply the vital signs to fatigue analysis by time- and frequency-domain. Despite the measuring results of the proposed indirect-contact ECG measurement are almost the same as the conventional ECG detection, the proposed one can be applied on the analysis of the HRV. The result of six subjects in the second experiment shows that the LF/HF of fatigue is larger than the ratio of freshness. Moreover, in the simulated driving experiment, the LF/HF value shows high consistency to the subject's self-judgement.