The relationship between some colors of cars and their image sensation are investigated by using the fuzzy set theory. The color image a person has of a car is quantified using the membership function. A computer program is constructed to calculate and analyze the similarity relations of various colors and further to arrange the corresponding membership functions of colors to one image word by size. The results show that the man-machine matching in color image is mostly dominated by the hue. The orderly table can be suggested to the designer for what color should be used if he can input the demanded image word (or words) for specifying the characteristic of a car in the stage of concept design. In such a way, not only can the error of a design decision be reduced, but the efficiency of the decision making can also be improved. Thus, it is indeed a powerful and valuable method for designing a product.