Facial Expression Recognition Based on FB2DPCA and Multi-classifier Fusion
Bin Hua, Ting Liu · 2009
A method of feature block two-dimensional principal component analysis (FB2DPCA) and multi-classifier combination is proposed for facial expression recognition. First, FB2DPCA is applied to extract human facial expression features, and then the expression classified result is obtained based on multi-classifier fusion with fuzzy integral. This proposed method is validated through the results of experiments on JAFFE facial expression database, and a high recognition rate is also achieved.