摘要
研究表明疲劳驾驶是引起交通事故的重要原因之一,因此有必要采取预防措施,而能够提前对事故进行准确预报并保证低误报警率是问题的关键所在.提出了利用多眼睑运动特征参数建立支持向量机模型进行疲劳预测的方法,其中眼睑运动特征参数是从驾驶模拟器上采集的眼电信号提取出的.根据Karolinska睡眠等级选出25名缺乏睡眠并在实验中撞到振动带的驾驶员,保证其开始驾驶阶段是警觉的,而事故发生阶段是疲劳的,然后将20名驾驶员作为训练对象,另5名驾驶员作为验证对象.结果表明,所用的方法可以提前至少5 min对由疲劳导致的事故进行预报.
Various investigations show that drivers＇ drowsiness is one of the main causes of traffic accidents. Thus, countermeasure device which should be able to predict the accidents accurately with low ratio of false alarms is currently required in many fields for sleepiness related hazard prevention. Drowsiness prediction was conducted by support vector machine （SVM） with eyelid related parameters extracted from the electrooculography（EOG） data collected in a driving simulator. 25 sleep-deprived subjects which hit the rumbles while driving in the experiment were selected based on the karolinska sleepiness scale （KSS） to make sure they were alert as they started driving and sleepy when the hits occurred, and then they were divided into training set including 20 subjects and validation set including the other 5 subjects. The validation results show that the hits can be successfully predicted at least five minutes ago by our SVM model.
出处
《北京航空航天大学学报》
EI
CAS
CSCD
北大核心
2009年第8期929-932,共4页
Journal of Beijing University of Aeronautics and Astronautics
基金
欧盟SENSATION资助项目FP6(507231)
关键词
支持向量机
眼电
疲劳预测
support vector machine
Electrooculography（EOG）
fatigue prediction
