Subpattern Complete Two Dimensional Locality Preserving Principal Component Analysis and its application to gait recognition

In this paper, a novel algorithm for feature extraction -Subpattern Complete Two Dimensional Locality Preserving Principal Component Analysis (SpC2DLPPCA) is proposed. The improved SpC2DLPPCA algorithm over C2DLPPCA and Subpattern Complete Two Dimensional Principal Component Analysis (SpC2DPCA) methods benefits greatly to three points: (1) SpC2DLPPCA can overcome a failing that larger dimension matrix may bring about more consuming time on computing its eigenvalues and eigenvectors. (2) SpC2DLPPCA can extract local information to implement recognition. (3)The idea of subblock is introduced into Two Dimensional Principal Component Analysis (2DPCA) and Two Dimensional Locality Preserving projections (2DLPP), so SpC2DLPPCA can preserve local neighbor graph structure and compact the expression of features. Finally, experiments on the CASIA(B) gait database show that SpC2DLPPCA has higher recognition accuracies than C2DLPPCA and SpC2DPCA.