由数码相机信号直接重构物体表面的光谱信息对不同照明条件下颜色的复现十分重要。基于普通商用数码相机分别摄取两种CIE标准光源D65和A下颜色样品的RGB信号, 采用主元分析法(PCA)和多项式模型的结合算法进行物体表面色光谱反射比的重构研究。实验结果表明, 同时应用两种光源下的RGB信号可有效提高光谱重构精度, 并能较好地再现色样的同色异谱特性。

Directly recovering the spectral information of surface colors using digital camera signals is important for color reproduction under various illuminating conditions. The RGB signals for color samples were obtained by a commercial digital camera under two different CIE illuminants, D65 and A, to reconstruct their corresponding spectral reflectance based on the algorithm combining principal component analysis(PCA) and polynomial model. The experimental results indicate that the spectral reconstruction accuracy could be improved by simultaneously utilizing the RGB signals under the two illuminants, in which case the metameric characteristics of color pairs could also be well reproduced.