摘要
研究了一种基于单导EEG的高空缺氧所致疲劳的实时检测技术。采用小波包分解对所采集的EEG进行预处理,然后选取EEG数据的近似熵和Welch谱分析中30-60Hz频段的能量作为模式识别的特征向量,用Bayes概率统计识别方法进行模式识别以检测高空缺氧所致疲劳。实验结果表明,通过提取的特征量可以很好的区分正常情况和缺氧疲劳情况下的脑电模式(检测正确率高于93%)。这种方法为客观、实时检测缺氧所致疲劳提供了可能。
This study suggests a real-time method to detect the fatigue due to high altitude hypoxia based on single-channel EEG. Based on the wavelet package decomposition, the EEG signal is pre-processed. Both the approximate entropy and the specific energy in a sub-band of 30-60 Hz of the Welch spectrum are extracted as the feature vector. The Bayesian classifier is utilized for pattern recognition to detect the fatigue due to high altitude hypoxia. Experimental results show that the EEG of the subjects with fatigue due to high altitude can be distinguished from normal one （the correction rate is higher than 93%）. This method may be used for objective real-time detection of the fatigue due to high altitude hypoxia.
出处
《传感技术学报》
CAS
CSCD
北大核心
2006年第4期1042-1044,共3页
Chinese Journal of Sensors and Actuators
关键词
缺氧
脑电EEG
模式识别
hypoxia, EEG
（electroence phalograhy electro encephalogralhy） pattern recognition
