Emotion recognition from physiological signals

Area of Science:

• Human-Computer Interaction
• Affective Computing
• Biomedical Signal Processing

Background:

• Emotion recognition is crucial for effective human-computer interaction.
• Current methods often struggle with accuracy and real-time application.
• Physiological signals offer a promising, objective modality for emotion detection.

Purpose of the Study:

• To develop and evaluate an emotion recognition approach using physiological signals.
• To analyze six basic emotions (joy, sadness, fear, disgust, neutrality, amusement).
• To improve the accuracy of emotion detection in human-computer interaction.

Main Methods:

• Emotions were induced using the International Affective Picture System (IAPS).
• Physiological signals measured included EMG, RV, SKT, SKC, BVP, and HR.
• Support Vector Machine (SVM) was employed for emotion classification based on extracted parameters.

Main Results:

• The proposed methodology achieved a general recognition rate of 85% across different emotional states.
• Characteristic parameters extracted from physiological signals effectively differentiated emotions.
• SVM classification demonstrated robust performance in identifying emotional states.

Conclusions:

• The developed approach based on physiological signals is effective for emotion recognition.
• This method holds significant potential for advancing human-computer interaction systems.
• Accurate emotion recognition via physiological data paves the way for more responsive and empathetic technology.