基於眼電圖法的疲勞駕駛偵測;Fatigue-Driving Detection Based on EOG

NCU Institutional Repository > 資訊電機學院 > 電機工程研究所 > 博碩士論文 > Item 987654321/74906

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/74906

题名:	基於眼電圖法的疲勞駕駛偵測;Fatigue-Driving Detection Based on EOG
作者:	張袁;Chang, Yuan
贡献者:	電機工程學系
关键词:	眼電圖;眼動訊號;疲勞;駕駛;瞌睡
日期:	2017-07-26
上传时间:	2017-10-27 16:11:33 (UTC+8)
出版者:	國立中央大學
摘要:	人們在開車時會因為開車時間過長，產生疲勞抑或想睡覺的情況。這些情形發生時，往往會造成嚴重的行車安全問題。近年來，疲勞駕駛事件頻傳，越來越多人投入這方面的研究，希望發展出一套警示系統，提醒駕駛人，降低車禍發生的機率。本研究以眼電圖法 (Electro-OculoGraphy, EOG)來擷取眼睛移動的軌跡，藉由測量眼睛周圍微弱的電位變化，得知眼部動作與特徵。本系統將 EOG 訊號擷取後，經由數位訊號處理來濾除雜訊，利用 EOG 水平訊號、EOG 水平微分訊號、 EOG 垂直訊號和 EOG 垂直微分訊號，獲得 EOG 的眨眼和掃視特徵，觀察眼動水平訊號的掃視速率，與眼動垂直訊號的眨眼頻率、閉眼持續時間的長短、閉眼後開眼的速率，作為判斷疲勞瞌睡的依據，辨識出受測駕駛者的駕駛狀態。本系統將受測駕駛者的狀態分成: 清醒、專注開車、疲勞與瞌睡四種狀態，經過 8 個志願者的測試，透過一款名為”歐洲卡車” 的遊戲進行模擬駕駛，讓受測駕駛者連續開車模擬 2 小時，比較一開始、一小時後、二小時後受測者的駕駛狀態。本系統可以偵測出當駕駛員在開車時，是否處於疲勞、瞌睡的狀態;當系統連續偵測到疲勞、瞌睡的訊號，將對駕駛員提出警示。未來，可以參考不同的生理訊號(EEG、ECG)，比較在疲勞、瞌睡時，與眼動訊號上的差異，修正系統的眼動參數閾值，讓系統在偵測上更為精準。 ;Driver fatigue and drowsiness are an important factor causing serious traffic accidents. In recent years, fatigue driving results in many people deaths or hurts. Therefore, more and more people put into these researches about how to detect driver fatigue and drowsiness. We hope to design a warning system for fatigue sleeping to reduce traffic accidents. In this study, we measured the eye movements by electro-oculography(EOG). The potential difference measured between the skins above and below the eyes reflected the eye movements in the vertical direction; that measured between the skins to the right and to the left of the eyes reflected the eye movements in the horizontal direction. We analyzed the saccade velocity of the EOG horizontal signal, the eye blinking frequency, the duration time of eye closure and the velocity of eye opened after closing the eyes. The fatigue warning system was developed on the basis of these analysis. The system was evaluated by emulation with a game named Euro Truck. The experiment was tested by 8 volunteers. The system can detect the fatigue and sleepy signals. When the signals were detected continuously, the system will warn the driver. In the future, we can compare with different physiological signals (EEG / ECG) and modify the eye movement parameters threshold so that the system will be more accurate in detection.
显示于类别:	[電機工程研究所] 博碩士論文

文件中的档案:

档案	描述	大小	格式	浏览次数
index.html		0Kb	HTML	193	检视/开启

在NCUIR中所有的数据项都受到原著作权保护.

社群 sharing

数据加载中.....