Digital image camera has received more and more attention because of its convenience in storing and transferring, the still exist problems about it are also hot topics of research. Auto white balance is one of the problems, it’s the result of differences between image sensors and human eyes. If the illumination of environment has changed, color cast will happen in image from sensors, but image from eyes due to color constancy won’t. For weakening this inconsistence and acquiring image of same scene under canonical illumination, color adjustment according to color temperature of environment should be considered. In this paper, an auto white balance approach combined gray world and coincidence of chromaticity histogram (GWCCH) is proposed. It’s based on basic assumptions of these two methods, measures color components in image, and selects appropriate routine and arguments to implement auto white balance. In the experiment results, the proposed method can meet the theory of gray world (GW) or coincidence of chromaticity histogram (CCH) respectively, and get good effect in more scenes than these two methods.