(2D)2PCA plus MMC: A new feature extraction for face recognition

In this paper, we combine the advantages of (2D)
²
PCA and MMC, and propose a two-stage framework: “(2D)
²
PCA+ MMC”. Since the extracted features based on (2D)
²
PCA are most expressive and based on maximal margin criterion (MMC) are robust, stable and efficient, in the first stage, a 2D two-directional feature extraction technique, (2D)
²
PCA, is employed to condense the dimension of image matrix; in the second stage, the linear discriminant analysis (MMC) is performed in the (2D)
²
PCA subspace to find the optimal discriminant feature vectors. In addition, the proposed method can make use of the descriptive information and discriminant information of the image. Experiments conducted on ORL and Yale face databases demonstrate the effectiveness and robustness of the proposed method.