FACIAL EXPRESSION RECOGNITION BASED ON FUSION OF GABOR AND LBP FEATURES

A new person-independent facial expression recognition algorithm based on fusion of Gabor and LBP Features with OLPP is presented. Aiming at the deficiency of common Gabor feature extraction method, a new Gabor feature extraction method is proposed. Each Gabor wavelet representation of an image is divided into small subblocks, then the mean value and standard deviation in each subblock are calculated, and the statistics of all Gabor wavelet representations are connected as the feature vector. Taking into account the outstanding performance of LBP to extract local texture, we combine local statistic features of Gabor wavelets with the LBP features as feature vector. After using OLPP to reduce the feature dimension of fusion features, the facial expression image is classified by nearest neighbor method. Experimental results on two databases indicate the proposed method has higher recognition rate compared with single feature method and other methods.