Find in Library

In this paper, the problem of choosing a set of most separable colors in a given environment is discussed. The proposed method models the process of generating theoretically computed best colors, printing of these colors through a color printer, and imaging the printed colors through a camera into an integrated framework. Thus, it provides a feasible way to generate practically best separable colors for a given environment with a set of given equipment. A real world application (robust color choice for small-size league RoboCup competition) is used as an example to illustrate the proposed method. Experimental results on this example show the competitiveness of the colors learned from our algorithm compared to the colors adopted by other teams which are chosen via an extensive trial and error process using standard color papers.