Color removal method preserving local contrast

A usual color-to-gray conversion only considers lightness components, and differences of colors are sometimes not reflected in monochrome images because three-dimension data which is RGB color components is converted in one-dimensional gray level. Gooch et al. have proposed a color removal method that considers the difference of colors which is named as the Color2Gray algorithm. Although the Colo2Gray algorithm can obtain good results in many cases, we can still improve it. In this paper, we propose a new color-to-gray conversion which attaches importance to the local contrast improvement. The proposed method evaluates the local contrast by computing the local variance of pixel values in an input color image and using it, high contrast monochrome image is obtained. Experiments show the effectiveness of the proposed method.