摘要
相比人脸和语音表情识别,生理信号能更客观地反映人们的情感状态.基于生理信号的情感计算研究越来越引起人们的关注.针对情感状态具有模糊特性,提出应用模糊理论从生理信号中选取特征向量进行分类识别.与其他分类方式相比,采用模糊分类识别研究更有助于其他学科的研究人员对情感发生机理的理解和深入研究.比较模糊理论与其他分类方式的识别结果,实验结果表明模糊识别效果总体与其他方式差距不大,但对不同的状态识别效果差距较小.
Compared with face and voice expression recognition,physiological signal can more objectively reflect the people＇s emotional state.Affective computing based on Physiological signals has drawn increasing attention.Emotional state is fuzzy,and fuzzy recognition of emotion states is proposed.Compared with other classification methods,fuzzy classification is more useful to other researchers in understanding the mechanism of the emotion recognition system.Comparison of fuzzy theory and other classification of recognition results,it is shown that the overall effect of fuzzy identification is the same with other methods,but the recognition gaps of the different states are much smaller.
出处
《四川师范大学学报（自然科学版）》
CAS
CSCD
北大核心
2010年第5期707-710,共4页
Journal of Sichuan Normal University（Natural Science）
基金
国家自然科学基金(60572143)资助项目
关键词
情感生理信号
模糊识别
模糊熵
情感识别
emotion physiological signal
fuzzy recognition
fuzzy entropy
emotional recognition
