Illumination invariant human face recognition: frequency or resonance?

In this paper we suggest the use of resonance based decomposition of images for illumination invariant face recognition. Although illumination is mostly considered as the low-frequency part of images, these low-frequency contents may possess low- and/or high-resonance nature. We first assume that an input image can be considered as a combination of illumination and reflectance. The images are then decomposed into low- and high-resonance components simultaneously. Because the energy distribution of subbands of resonance based decomposition are different for an image with good illumination effects and an image with high illumination variations, the energy of subbands of the two components can be thresholded to deactivate the subbands with unwanted energy distribution created by illumination effects. For dimensionality reduction and classification the principal component analysis and extreme learning machine have been used, respectively. Experiments and comparisons illustrate the effectiveness of the proposed resonance based method in illumination invariant face recognition.