Saliency Enhanced Decolorization

Color-to-grayscale conversion methods are widely used in the field of computer vision and image processing. The importance of preserving the source color image details is really high, because only correctly converted images can be used in real world applications (e.g. a colorblind version of a webpage, etc.). The lack of some image details can cause misunderstandings and incorrect data evaluation. Saliency maps help us identify areas of the image that attract the visual attention of users. In this paper we propose a new method for color-to-grayscale conversion based on preserving image saliency.