Color Gamut Mapping Based on Image Fusion.

Yu Wang,Bin Li
DOI: https://doi.org/10.1109/csse.2008.369
2008-01-01
Abstract:Gamut mapping is a color transformation technique to solve a problem caused by mismatch of gamuts among imaging devices. One plausible goal of gamut mapping is to find a reproduction that is perceptually closest to the corresponding original image when an exact color matching is not possible. Most of the classical gamut mapping methods use colorimetric pixel-wise principles and ignore the spatial color configuration. The recently proposed spatial-dependent approaches are either based on heuristic assumptions or involve a high computational cost. In this paper, a new approach for image-dependent gamut mapping via image fusion is proposed. First, a measure to quantify the perceptual difference between images is presented, with which a divide and conquer scheme is employed to take image colors and spatial characteristics into account. Then a mapping fusion model is constructed to implement the method. Experiments show that the proposed algorithm gains obvious advantage than traditional ones, especially in color difference and image details.
What problem does this paper attempt to address?