Research of emotion recognition based on pulse signal

As one of the most important human physiological signals, pulse signal contains a wealth of information which can reflect the change of emotions. In this paper, time-domain features and wavelet coefficient features through wavelet transform are extracted respectively from raw pulse signals. Then the improved Max-Min Ant System (IMMAS) combined with Fisher classifier is used for feature selection and the classification of happiness, disgust, grief, anger, fear. Finally, not only good recognition effect, but also some useful feature combinations are obtained.