According to the present embodiment, the first relational equations which represent the relationships between the spectral reflectances of the maximum gradation values and the spectral reflectances of the characteristics-acquired gradation value (the intermediate gradation value for which the spectral reflectances are obtained regarding the prediction target color) are obtained for each of the sample colors, and using the first relational equations for respective sample colors, the prediction values of the spectral reflectances of the characteristics-acquired gradation value for the prediction target color are obtained. Then, the difference values between the prediction values and the actual measurement values are obtained, and a sample color for which the minimum difference value is obtained is selected as the reference color. The second relational equation that represents the relationship between the spectral reflectances of the maximum gradation value and the spectral reflectances of the prediction target gradation value for the reference color is obtained, and the spectral reflectance of the maximum gradation values for the prediction target color are applied to the second relational equation, whereby the spectral reflectances (prediction values) of the prediction target gradation value for the prediction target color are obtained. As above, the spectral reflectances are predicted for the prediction target color using, as the reference color, the color in which known spectral reflectances for the prediction target color can be predicted with highest accuracy. Accordingly, highly accurate prediction values are obtained. Thus, according to the present embodiment, in a case in which there is an intermediate gradation value for which the spectral reflectances are known regarding the prediction target color, high accuracy prediction of the spectral reflectances of the prediction target gradation values for the prediction target color is enabled using the information on the known spectral reflectances.