Face expression recognition based on equable principal component analysis and linear regression classification
Yani Zhu, Xiaoxin Li, Guohua Wu · 2016
In this paper, we present a novel expression recognition method. Equable Principal Component Analysis (EPCA) is used as expression features representation and Linear Regression Classification (LRC) is employed as expression classifiers. EPCA maintains the useful information of the original image while reducing the dimension of feature vector data. LRC deals with the problem of face recognition as a linear regression problem. Experiments of human images are performed on the Yale and JAFFE database. Compared to the state-of-art approaches, the recognition rate of the proposed method is higher. Therefore, the combination of LRC and EPCA for facial expression recognition is feasible.