Compensation of de-saturation effect in HDR images based on real scene adaptation analysis

Chromatic adaptation transform is used to many color appearance models. Also, the iCAM06 model, a high-dynamic range (HDR) rendering process contains chromatic adaptation and is based on the CIECAM02 model. The D factor or degree of adaptation in chromatic adaptation of the iCAM06 model is the function of surround luminance or adapting luminance level, but does not consider visual white point shift according to the variation of luminance level. The iCAM06 model acquires a white image from the Gaussian low-passed image. So, each pixel has different white value, and when some of colors cover the whole image, de-saturation effect appears around these colors. In addition, the scale factor used to reduce de-saturation effect causes unsuitable change of adaptation degree. Therefore, we propose the new method to find global illuminant information and degree of adaptation in the HDR image rendering process. By comparing the proposed with the conventional method, we confirm that the proposed method has better performance in illuminant estimation and color representation.