脑电信号中独立分量特征提取与脑力负荷分类  

Extraction of Independent Component Features in Electroencephalogram Signals and Classification of Mental Workload

作　　者：曲洪权[1] 单一平 刘欲哲 庞丽萍[2] 范占利 王浚[2] QU Hong-quan;SHAN Yi-ping;LIU Yu-zhe;PANG Li-ping;FAN Zhan-li;WANG Jun(Information College,North China University of Technology,Beijing 100144,China;College of Aviation Science and Engineering,Beihang University,Beijing 100191,China)

机构地区：[1]北方工业大学信息学院,北京100144 [2]北京航空航天大学航空科学与工程学院,北京100191

出　　处：《科学技术与工程》2020年第28期11499-11504,共6页Science Technology and Engineering

基　　金：国家自然科学基金(XLYC1802092)。

摘　　要：脑力负荷过高会造成作业绩效下降和人因事故,过低则会造成人力资源浪费,所以研究操作人员脑力负荷状态非常有意义。现有脑力负荷分类方法利用脑电(electroencephalogram,EEG)信号特征进行分类,准确率较低。所以,针对视觉和操作类脑力负荷提出一种基于脑电独立分量特征的分类方法,该方法采用独立分量分析(independent component analysis,ICA)对脑电信号进行分离,直接对得到的独立分量提取四种不同频段的能量特征,最后将特征作为支持向量机(support vector machine,SVM)的输入,对脑力负荷进行分类。由于直接使用脑电独立分量特征,所以分类精度高于通用方法,平均分类精度提高29.14%。还进一步发现脑电独立分量中存在的眼电伪迹对分类结果没有明显影响。提出的方法可以实现快速、准确、自动的脑力负荷分类。The excessive mental workload will reduce work efficiency or even cause accident,but low mental workload will cause a waste of human resources.It is very significant to study the mental workload status of operators.The existing mental workload classification method is based on the features of electroencephalogram(EEG)signals,but its accuracy is low.So,a mental workload classification method based on features of EEG independent components for vision and operation was proposed,which was implemented by the following four steps of filtering,obtaining EEG independent components,extracting independent components energy features and classifying.This method used independent component analysis(ICA)to separate EEG signals and directly extract the obtained independent components.The energy features of four different frequency bands were used as input to a support vector machine(SVM)to classify the mental load.Since this method directly used independent components energy features for feature extraction,it could obtain better classification results.Compared with the general method,the average accuracy is increased by 29.14%.Further,it was found that the Ocular Artifacts have no significant effect on the classification of mental workload based on features of EEG independent components.The presented method might provide a way to realize a fast,accurate and automatic mental workload classification.

关 键 词：脑力负荷 独立分量分析 支持向量机 脑电 

分 类 号：R338[医药卫生—人体生理学]