System for analysis of involuntary facial expressions

Facial expression is a rich source for ascertaining the emotional and psychological states of human In order to study facial expressions, one needs to accurately measure the degree of change official expression. In this study, we propose a system for measuring involuntary facial expression from time-sequential images obtained in watching video clips. This system measures seven parameters of facial components in three areas, eyebrows, eyes, and mouth. This system helps to detect facial expression changes. In an experiment, four subjects watched three video clips that would evoke some facial expression (e.g., surprise and laughter, disgust and so on). The frontal facial expressions of these subjects were recorded on the videotape using equipment that includes a half mirror. By measuring seven facial features from time-sequential images, two principal component factors were obtained. As a result, we were able to plot dynamically the facial expression sequence (neutral face, disgusted, surprised, and smile).