(2D)2 MMDA: a New Method of Feature Extraction and Face Recognition Based on Image Matrix

摘要

Based on maximum margin criterion, a new feature extraction method is proposed for face image recognition in the paper. The method is called 2-directional 2-dimensional maximum margin discriminant analysis ((2D)2 MMDA), which works image matrix in the row direction and in the column direction simultaneously for feature extraction. The experimental results on ORL face databases and YALE face databases indicate that the proposed method has the advantage of higher recognition rate, less memory requirements and better computing performance than the 2D-PCA and (2D)2PCA method.