Facial Expression Recognition in Various Internal States Using Independent Component Analysis

摘要

This paper presents a new approach method to recognize facial expressions in various internal states using independent component analysis (ICA). We developed a representation of facial expression images based on independent component analysis for feature extraction of facial expressions. This representation consists of two steps. In the first step, we present a representation based on principal component analysis (PCA) excluded the first 2 principal components to reflect well the changes in facial expressions. Second, ICA representation from this PCA representation was developed. Finally, classification of facial expressions in various internal states was created on two dimensional structure of emotion with pleasure/displeasure dimension and arousal/sleep dimension. The proposed algorithm demonstrates the ability to discriminate the changes of facial expressions in various internal states. This system is possible to use in cognitive processes, social interaction and behavioral investigations of emotion.