A Fusion of Face Symmetry of Two-Dimensional Principal Component Analysis and Face Recognition
Fulong Wang, Cheng Huang, Xiaoliang Liu · 2009
In this paper, a fusion of facial symmetry information method is developed for improving two-dimensional principal component analysis. The proposed method uses the characteristic of facial symmetry to generate odd-even symmetry images, by weighting odd-even symmetry matrix to replace original image matrix extracting features, and at last minimum distance classifier is used for classification. The predominance of this method is that it takes full advantages of facial symmetry information and considers the impact of odd symmetry matrix which reflects the non-symmetric in Face Recognition. The experiment results on the YALE and ORL face database show that this method has better performance and robustness than the classical PCA and 2DPCA.