Application of 2DPCA Based Techniques in DCT Domainfor Face Recognition

摘要

In this paper, we introduce 2DPCA, DiaPCA and DiaPCA+2DPCA in DCT domainfor the aim of face recognition. The 2D DCT transform has been used as a preprocessing step,then 2DPCA, DiaPCA and DiaPCA+2DPCA are applied on the upper left corner block of theglobal 2D DCT transform matrix of the original images. The ORL face database is used tocompare the proposed approach with the conventional ones without DCT under Four matrixsimilarity measures: Frobenuis, Yang, Assembled Matrix Distance (AMD) and Volume Meas- Measure(VM). The experiments show that in addition to ure the significant gain in both the training andtesting times, the recognition rate using 2DPC 2DPCA, DiaPCA and DiaPCA+2DPCA in DCT do- A, domainis generally better or at least competitive with the recognition rates obtained by applyingmain these three 2D appearance based statistical techniques directly on the raw pixel images; especiallyunder the VM similarity measure.