In everyday interaction, our face is the basic and primary focus of attention. Out of many human psycho-signatures, the face provides a unique identification of a person by the virtue of its size, shape, and different expressions such as happy, sad, disgust, surprise, fear, anger, neutral, etc. In a human computer interaction, facial expression recognition is an interesting and one of the most challenging research areas. In the proposed work, principle component analysis (PCA) and independent component analysis (ICA) are used for the facial expressions recognition. Euclidean distance classifier and cosine similarity measure are used as the cost function for testing and verification of the images. Japanese Female Facial Expression (JAFFE) database and our own customized database are used for the analysis. The experimental result shows that ICA provides improved facial expression recognition in comparison with PCA. The PCA and ICA provides detection accuracy of 81.42 and 94.28%, respectively.
CITATION STYLE
Patil, M. N., Iyer, B., & Arya, R. (2016). Performance evaluation of PCA and ICA algorithm for facial expression recognition application. In Advances in Intelligent Systems and Computing (Vol. 436, pp. 965–976). Springer Verlag. https://doi.org/10.1007/978-981-10-0448-3_81
Mendeley helps you to discover research relevant for your work.