Discriminative Locality Preserving Canonical Correlation Analysis

In this paper, we propose a novel dimension reduction method based on canonical correlation analysis, called discriminative locality preserving canonical correlation analysis (DLPCCA) method. In particular, we use discriminative information to maximize the correlations between intra-class samples, and maximize the margins between inter-class samples. Moreover. local preserving data structure can be used to estimate the data structure, and thus DLPCCA achieves better performance. Experimental results on Yale and ORL datasets show that DLPCCA outperforms the representative algorithms including CCA, KCCA, LPCCA and LDCCA.