提出了一种基于色域分析与聚类分析结合的基色筛选算法。该方法对传统4色色域和目标高保真色域进行分析，筛选出4色色域外的特征点，并对其光谱反射率进行主成分分析，确定基色的个数m。以m作为分类数，对集合中特征色光谱反射率进行聚类分析，在m个分类光谱中选择和特征色平均光谱距离最远的m个光谱作为基色光谱，在基色库中选择和m个光谱相关距离最小的实际专色作为最终基色。实验结果表明，提出的基色选择算法能够准确估计出新基色的光谱反射率，最大程度实现了目标高保真色域的再现。

A method of basis colorant selection based on gamut analysis and cluster analysis is proposed. Conventional 4-color gamut and target hifi-color gamut are analyzed to pick out the featured color sets where the colorants exist, and the principal component analysis is performed for the spectral reflectance of these colors to get the numbers of basis colorant m. With the cluster number m, spectral reflectance of the featured colors in the set is divided by clustering method, and the average spectral reflectance of the featured colors is calculated. m spectral reflectance among the m cluster which are most nearest to the average spectral reflectance is selected as the spectral reflectance of the basis colorant. The results show that the proposed method can estimate spectral reflectance of new basis colorant accurately, which realises the maximun reproduction of the target high fidelity color gamut.