Reflection Components Separation using Chromaticity and Noise Analysis

A method to separate reflection components of uniformly colored surfaces using a single image is presented. The method is based mainly on chromaticity, particularly on the distribution of specular and diffuse points in a two-dimensional chromaticity intensity space. We found that by utilizing the space, the problem of reflection components separation can be simplified into the problem of identifying diffuse chromaticity, i.e., by knowing correct diffuse chromaticity of a surface, the separation can be done straightforwardly and accurately. In order to find correct diffuse chromaticity, an analysis on noise is required, since most real images are suffered from it. Unlike existing methods, the proposed method is able to identify diffuse chromaticity robustly for any kind of surface roughness and light directions, without requiring diffuse-specular pixels segmentation. In addition, to obtain pure white specular component, we present a handy normalization technique that does not require approximated linear basis functions. The experimental results show that the proposed method is accurate and robust for real images under a single colored scene illumination.