Color Constancy for Multiple-Illuminant Scenes using Retinex and SVR

Scenes lit by multiple colors of illumination provide a problem for color constancy and automatic white balancing algorithms. Many of these algorithms estimate a single illuminant color, but since when there are multiple illuminants, there is in fact not a single correct answer. For automatic white balancing and color-cast removal in digital images, multiple illuminants mean that a single, image-wide adjustment of colors may not yield a good result, since the adjustment that makes one image area look better, may simultaneously make another look worse. Retinex is one method that adjusts colors on a pixel-by-pixel basis, and so inherently addresses the multiple-illumination problem, but it does not always produce a perfect overall color balance. On the other hand, illumination estimation by Support Vector Regression (SVR), produces quite good overall color balance for single-illuminant scenes, but does not adjust the colors locally. By combining Retinex and SVR in to a hybrid Retinex+SVR method, some of these problems can be overcome. Experiments with both synthetic and real images show promising results.