一种基于核主成分分析的驾驶疲劳状态预测方法  

A method for predicting driving fatigue state based on kernel principal component analysis

作　　者：郑超英 叶贝鸽 刘萍[2] 邱桃荣[2] ZHENG Chaoying;YE Beige;LIU Ping;QIU Taorong(College of Engineering,Huaqiao University,Quanzhou,362021,China;School of Information Engineering,Nanchang University,Nanchang 330031,China)

机构地区：[1]华侨大学工学院,福建泉州362021 [2]南昌大学信息工程学院,江西南昌330031

出　　处：《南昌大学学报（理科版）》2019年第2期182-186,共5页Journal of Nanchang University(Natural Science)

基　　金：国家自然科学基金资助项目(61762045,81460769,61070139)

摘　　要：针对在驾驶疲劳状态识别研究中采集的脑电信号(Electroencephalograph,EEG)具有非线性特点,以及目前基于EEG的驾驶疲劳状态识别方法仍存在识别准确率不很理想的问题,提出基于样本熵和核主成分分析的驾驶疲劳状态识别方法。所研究的方法结合了基于样本熵在驾驶疲劳状态识别方法具有较好的识别准确率优势以及核主成分对主成分是非线性的情况下具有较好的降维效果优势。实验结果表明所提出的方法是有效的。The EEG data acquired in the study of driving fatigue state recognition has nonlinear characteristics,and the current recognition method based on EEG for driver fatigue state still has the problem that the recognition accuracy rate is not ideal.This paper proposes a driving fatigue identification method based on the sample entropy(SE)and the kernel principal component analysis(KPCA).The proposed method combines the advantages better recognition accuracy of SE in the recognition of driving fatigue state and the benefits of satisfactory dimensionality reduction of KPCA.The experimental results show that the proposed method is effective.

关 键 词：驾驶疲劳 样本熵 核主成分分析 支持向量机 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]