The appearance of a face will vary dramatically when the illumination changes

which makes face recognition even more challenging. In this paper

an illumination compensation method by means of multi-level wavelet decomposition and spline interpolation is proposed to reduce the effects of illumination variations on face recognition. A lighting change in an image can be modeled as an additive noise in addition to a multiplicative one. The additive noise

called background

can be estimated and taken out by multi-level wavelet decomposition followed by spline interpolation; the multiplicative one

called gain

can be done as the similar way in the logarithm domain. Experimental results on Yale B face database show that the proposed method achieves 2~11% recognition rate upper than the others. It can efficiently composite the facial images under varying illumination conditions

as well as improve the recognition rate and robustness of face recognition.