论文信息 - Face recognition based on 2DPCA, DIAPCA and DIA2DPCA in DCT domain

Face recognition based on 2DPCA, DIAPCA and DIA2DPCA in DCT domain

In this paper, we introduce 2DPCA, DiaPCA and DiaPCA+2DPCA in DCT domain for the aim of face recognition. The 2D DCT transform has been used as a preprocessing step, then 2DPCA, DiaPCA and DiaPCA+2DPCA are applied to a wtimesw upper left block of the global 2D DCT transform matrix of the original images. The experiments which are performed on the ORL face database show that: in addition to the expected considerable gain in both the training and testing time, the recognition rate is higher using 2DPCA, DiaPCA and DiaPCA+2DPCA in DCT domain than applying these three methods directly on the raw pixel images.

[1] Martin D. Levine,et al. Face Recognition Using the Discrete Cosine Transform , 2001, International Journal of Computer Vision.

[2] M. Turk,et al. Eigenfaces for Recognition , 1991, Journal of Cognitive Neuroscience.

[3] Alejandro F. Frangi,et al. Two-dimensional PCA: a new approach to appearance-based face representation and recognition , 2004 .

[4] Jian Yang,et al. From image vector to matrix: a straightforward image projection technique - IMPCA vs. PCA , 2002, Pattern Recognit..

[5] Conrad Sanderson,et al. On Local Features for Face Verification , 2004 .

[6] Daoqiang Zhang,et al. Diagonal principal component analysis for face recognition , 2006, Pattern Recognit..

[7] Meng Joo Er,et al. PCA and LDA in DCT domain , 2005, Pattern Recognit. Lett..