基于脑电信号的驾驶员情绪状态识别研究

为研究驾驶员情绪状态识别技术,利用特定题材的影音氛围诱发兴奋和悲伤2种情绪,结合汽车模拟驾驶试验采集驾驶员情绪化驾驶时的脑电信号,利用db5小波分解算法分解信号,提取受情绪状态显著的脑电信号成分,并计算相应的功率谱。结果表明:驾驶员左右额叶区脑电β波受情绪状态影响显著;平静状态时,驾驶员左右额叶脑电β波功率值Pβ大小近似,数值较小;兴奋状态时,左右额叶脑电Pβ大小近似,相比于平静时显著增加;悲伤状态时,左额叶脑电Pβ显著增加,右额叶脑电Pβ相比于平静时无显著变化。Pβ可作为区分驾驶员平静(平常)、兴奋和悲伤这3种情绪状态的依据。

Study on an Emotional State Recognition Technology Based on Drivers' EEGs

In order to study an emotional states recognition technology,the exciting and sad emotions of drivers were induced by some certain contexts of video atmospheres.EEGs of the drivers were acquired with the help of a vehicle simulation machine while driving.The db5 wavelet algorithm was used to decompose the EEGs.Common components of EEGs which have significant changes under different emotional states were extracted out.Power spectrums of the signals were worked out.Results show that β waves of the EEGs from both left and right frontal lobes have significant changes in different emotional states commonly.When the drivers are in calm,powers of the β waves from left frontal lobes are almost same as the ones from the right frontal lobes with small values.When the drivers are excited,powers of the β waves from both left and right frontal lobes are close,but much bigger than the corresponding ones when the drivers are in calm.When the drivers are sad,powers of the β waves become significantly bigger while the ones from right frontal lobes remain almost the same,compared with the corresponding ones when the drivers are in calm.Emotional states such as calm(or normal),excitement and sadness can be recognized effectively according to the powers of β waves.