Emotion visualization system based on physiological signals combined with the picture and scene

Abstract

In this paper, the system of emotion visualization and the system of emotion recognition and judgment are established. Twenty subjects were selected for the test on the above two systems, meanwhile the emotional trend changes given by emotion judgment system based on the set of optimal signal feature and based on conventional machine learning-based method are compared. The results show that the emotional trend changes given by emotion visualization based on picture and scene change are roughly consistent with those obtained by emotion judgment system. As to the real-time ability and interactivity of emotional judgment, the emotion visualization based on scene change is better than that based on picture change; the emotion judgment system based on the set of optimal signal feature is better than the system based on the conventional machine learning-based method. The test experience of subjects has an impact on the test results. Multi-dimensional interactive environment is easier to affect people’s emotional changes than single dimensional interactive environment.