基于脑功能网络的脑疲劳状态检测研究

为了对驾驶员的疲劳状态进行有效识别,进行了真实高速公路驾驶实验。通过无线脑电采集设备采集驾驶员在不同时刻的多导脑电数据。基于相位滞后指数对不同时刻的脑电数据分别建立邻接矩阵、二值矩阵,并构建脑功能网络,绘制脑网络地形图。利用复杂网络理论计算和分析脑功能网络节点特征参数--度,并对不同时刻的节点度进行对比。根据各个节点度的变化趋势以及驾驶员的主观判断,发现随着驾驶实验的继续,驾驶疲劳程度加深,大脑的信息处理能力降低,相位滞后指数值与节点度下降趋势明显。表明脑功能网络特征参数--度,能够作为表征大脑疲劳的客观指标。通过与其他检测方法的对比得到采用节点度作为脑疲劳状态评价指标可靠性更强。

Research on Fatigue Driving Recognition Based on Brain Function Network

To effectively identify the drivers fatigue status, a real highway driving experiment is conducted. Collecting multi-channel electroencephalograph (EEG) of drivers at different times through wireless EEG acquisition equipment. Based on the phase lag index, an adjacency matrix and a binary matrix are established. A brain function network and its corresponding topographic map are obtained. The complex network theory is used to calculate and analyze the characteristic parameters of the brain function network node-degree and compare the node degree at different moments. According to the variation trend of each node degree and the driver's subjective judgment, it is found that as the driving experiment continues, the degree of driving fatigue increases, the information processing ability of the brain decreases, and the decrease trend of phase lag index value and the node degree is obvious. It shows that the degree of brain function network can be used as an objective index of brain fatigue. Through comparison with other detection methods, it is obtained that using the node degree as the evaluation index of brain fatigue is more reliable.