This paper provides a novel color balance algorithm based on fuzzy inference rules by basic concept of color and the experimental results. In this algorithm, the color sample database is forms with the images are captured in different illuminants at first. Second, statistical results of color sample database are used to form fuzzy inference system. Finally, image is adjusted from unknown illuminant to the image in standard illuminant with correction the fuzzy inference system forecast. Experimental results and a comparative study justify that the proposed method is preferable to existing methods with regard to execution time and the performance.