Abstract: Color night vision technology is the current forefront in night vision technology field. How to achieve an objective evaluation of color image quality for night vision is still a problem. A new subjective visual evaluation experiment on color harmony was designed and conducted in accordance with the international standards for the underrepresentation problem of the last experiment pointed out by Gao Shaoshu, a BIT doctor. The color fusion images tested for subjective evaluation in the new experiment were divided into three types of typical sceneries: green plants, sea to sky, and urban buildings; experimental subjects were increased to 52. The regression analysis of the experiment results shows that in the three types of sceneries, the assessment model proposed by Gao Shaoshu is in relatively good consistency with the subjective evaluation results, and can be used to make an effective objective evaluation of color fusion images' color harmony in different sceneries.