基于前额单通道脑电分析的睁闭眼状态检测  

DETECTION OF OPEN AND CLOSED EYES BASED ON SINGLE-CHANNEL EEG ANALYSIS OF FOREHEAD

作　　者：吴昊 张建海[2] Wu Hao;Zhang Jianhai(Zhuoyue Honors College,Hangzhou Dianzi University,Hangzhou 310018,Zhejiang,China;School of Computer Science and Technology,Hangzhou Dianzi University,Hangzhou 310018,Zhejiang,China)

机构地区：[1]杭州电子科技大学卓越学院,浙江杭州310018 [2]杭州电子科技大学计算机学院,浙江杭州310018

出　　处：《计算机应用与软件》2023年第7期50-54,共5页Computer Applications and Software

基　　金：国家自然科学基金重点项目(61633010)。

摘　　要：眨眼动作变缓、闭眼持续时间增加作为人困倦时的显著特征,可以作为对疲劳驾驶状态判别的准确指标。已有的基于眼肌电的眨眼检测无法有效识别缓慢眨眼情况。为了解决缓慢眨眼识别的问题,提出一种基于前额单通道脑电数据分析的方法。采集睁闭眼状态下脑电信号;提取脑电信号特征,特别是α波段信号的时频变化特征;采用SVM分类算法,实现睁眼和闭眼状态的快速准确检测,准确率可达87.9%。该方法与已成熟的基于前额肌电的面部状态识别(眨眼频率、打哈欠等)相结合,可以显著提高疲劳状态检测的有效性和可靠性,具有重要的实用价值。The slower blinking action and the increased duration of closed eyes are the salient features of people when they are drowsy,and can be used as accurate indicators for judging the fatigued driving state.The existing blink detection based on EOG or EMG cannot effectively identify slow blinking.In order to solve the problem of slow blink recognition,a method based on forehead single-channel EEG data analysis is proposed.The EEG signals were collected in the state of open and closed eyes,and the EEG signal characteristics were extracted,especially the time-frequency change characteristics of the alpha band signal.The SVM classification algorithm was used to realize the fast and accurate detection of the open and closed eyes state,and the accuracy rate could be up to 87.9%.This method was combined with mature facial state recognition based on forehead EMG(blink frequency,yawning,etc.),which could significantly improve the effectiveness and reliability of fatigue state detection,and had important practical value.

关 键 词：脑电信号 疲劳检测 缓慢眨眼检测 特征提取 Α波 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]