Thermal imaging based affective state recognition

摘要

Human Robot Interaction is a multidisciplinary field which involves developing, perceiving and assessing robotic systems. In effective communication, the understanding of emotions and intentions are essential. A robotic system that is capable of recognizing emotion states and synthesizing proper response would be beneficial for Human Robot Interaction (HRI). Human emotion recognition could be achieved through affective state computing method. Emotional state assessment of humans has been traditionally studied using various direct physiological measures and psychological self-reports. There are several measures to identify human emotional states such as gestures, facial images, physiological signals and neuro-imaging methods. However, some of these approaches require sizeable and expensive equipment which might hinder free movement. Recent finding shows that facial cutaneous temperature and its topographic distribution exhibited specific features clearly correlated to emotional arousal and associated measures of standard physiological signals of the sympathetic activity. The results of this study indicate thermal imaging as an alternative, contactless and non-invasive method for assessing individual's emotional arousal in psychophysiology.