Pressure in long term may cause some diseases.It is important to assess the state of emotional stress reasonably and effectively.It is a reasonable method of pressure condition assessment based on the EEG (electroencephalograph),because of EEG contains plenty of emotional information.In this paper,focusing on the EEG signal characters extracting of emotional stress,an algorithm was investigated.The degree of random was quantified by the Kc factor.The complexity and energy distributing were quantfied by the approximate entropy and wavelet entropy.The Kc factor,approximate entropy and wavelet entropy were fused as the emotional characters by the optimal support vector machine.Based on principle of the overall optimization and the survival of the fittest of genetic algorithm,selection-crossover-mutation were done to pursue the optical parameters of SVM."Fruit Ninja" game was selected as a source of stress,and a total of 92 groups of EEG signals were collected from 8 subjects.Assessment results showed that the highest classification accuracy was 94.12％,and average accuracy was 82.06％.The level of sensitivity to the stress was different among the brain regions.The left hemisphere was more sensitive to stress than the right one.The research is expected to be helpful for people to take proper methods of relieving stress and restoring physical and mental health.%长期处于压力状态容易引发各类疾病,合理有效的情感压力状态评估是进行压力情感干预的基础.脑电数据含有丰富的情感信息,针对脑电数据的压力情感特征提取问题,提出一种基于Kc复杂度、小波熵与近似熵相结合的脑电数据的压力情感特征提取方法,以Kc复杂度因子来量化脑电数据的随机程度,以小波熵和近似熵参数分别在时域和频域来量化脑电数据的复杂程度与能量分布；采用遗传算法进行全局寻优,按适者生存的原则进行支持向量机参数的选择、交叉、变异,以此优化的支持向量机融合3类不同层次的特征参数,实现压力情感状态评估.以“切水果”游戏作为压力源,采集8名被试共92组脑电信号,基于该算法来分析被试者的压力状态,最高识别率为94.12％,平均识别率82.06％.研究表明,不同脑区对压力敏感程度不同,左半球相对右半球来说,压力感受敏感.希望通过此工作,可以帮助人们有针对性地采取相应措施缓解压力,恢复身心健康.
展开▼
