Illumination Invariant Using DT-CWT and EMD for Face Recognition
Yuehui Sun · 2008
In this paper we describe a method to address illumination isolation for face recognition with empirical mode decomposition (EMD) to decompose subimages of dual-tree complex wavelet transform (DT-CWT) firstl. After that, the subimages are reconstructed without illumination distortion components. We then perform face recognition using improved orthogonal neighborhood preserving projections (IONPP). Experimental results reported on the CMU PIE database have shown method proposed outperfoms well.