基于EEG的驾驶疲劳识别算法及其有效性验证

发布时间：2018-06-04 19:16

本文选题：驾驶疲劳 + 核主元分析　；参考：《北京工业大学学报》2017年06期

【摘要】：为有效识别驾驶员疲劳状态,基于脑电信号(electroencephalogram,EEG)提出了一种驾驶疲劳状态识别方法.首先,以时间段划分疲劳等级,并采用主、客观测评指标对疲劳等级划分的合理性进行验证.然后,利用快速傅里叶变换对脑电信号进行分析,在此基础上选取3种频段的平均幅值和5项合成指标,通过核主元分析(kernel principal component analysis,KPCA)构建疲劳识别脑电指标,结合支持向量机(support vector machine,SVM),构建了驾驶员疲劳状态识别模型.最后,采用30名驾驶员连续驾驶2 h的脑电数据,对该模型方法进行试算.试算结果表明:疲劳状态识别正确率为79.17%~92.03%,平均正确率为84.62%,该方法可用于驾驶疲劳识别.
[Abstract]:In order to identify driver fatigue state effectively, a driving fatigue state recognition method based on EEG electroencephalogramma (EGG) is proposed. First, the fatigue grade is divided by time, and the rationality of fatigue grade is verified by subjective and objective indexes. Then, the EEG signal is analyzed by using fast Fourier transform. On the basis of this, the average amplitude of three frequency bands and five synthetic indexes are selected, and the fatigue identification EEG index is constructed by kernel principal component analysis (kernel principal component analysis) and kernel principal component analysis (KPCA). Combined with support vector machine (SVM), a driver fatigue state recognition model is constructed. Finally, the EEG data of 30 drivers driving for 2 hours were used to calculate the model. The experimental results show that the correct rate of fatigue state recognition is 79.17 and 92.03, and the average correct rate is 84.622.This method can be used for driving fatigue recognition.
【作者单位】：西南交通大学交通运输与物流学院;
【基金】：国家自然科学基金资助项目(51108390) 国家重点研发计划资助课题(2016YFC0802209)
【分类号】：R318;TN911.7

【相似文献】