Facial Emotion Recognition Using Gravitational Search Algorithm

Facial expressions are the mirror of human’s internal emotions. They are one of the major ways to communicate with others. Thus, play a critical role in line with human machine communications. This paper proposes a new facial emotion recognition method based on Gravitational Search Algorithm, from eyes and mouth features. Detection is implemented by part-based appearance manners and these attributes are used in the recognition phase. In this phase, fa-cial attributes and their mapping to emotion space are encoded using GSA-based optimization. Applying the method on Ebner’s facial expression database indicates 69.43% accuracy.