Design and Implementation of a Real-time Emotion Recognition System Based on Physiological signals

In this research, we designed and implemented a real-time emotion recognition system by using physiological signals. The video materials were used to stimulate subjects' emotions of happiness, surprise, sadness, anger, fear and calmness. The MP160 physiological recorder was used to collect the physiological signals of ECG, myoelectricity, skin conductance, respiration and skin temperature of the subjects under the each emotion condition. After preprocessing, the combined algorithm including PCA and SVM was used to realize real-time classification of emotions. Finally, the system took four students as the subjects of the experiments, and the average recognition rate of the six emotions was 70%.