彩色扫描仪的光谱反射率估计方法

为从扫描仪获取精确的、设备无关的颜色描述，必须对扫描仪进行颜色表征(Characterization)。考虑到实际扫描仪的成像过程往往偏离线性反射模型，并且训练样本在颜色空间中的不同分布对颜色估计的精度有较大影响，提出了一种基于有限样本加权训练的彩色扫描仪光谱表征方法，从而可以从低维的RGB颜色值计算得到精确的光谱反射率。 实验结果表明，该方法在色度及反射率精度方面均明显优于先前的方法。

Estimation of Spectral Reflectance Using Color Scanner

To obtain accurate and device-independent color, the scanner needs to be characterized. Considering that the behavior of color scanner always deviates from linear reflectance model and that the color distribution of training samples has considerable effect on the color estimation accuracy, a new spectral characterization method is proposed. This method can estimate accurate high-dimensional spectral reflectance from low-dimensional scanner responses, based on limited sample selection and weighted training. Experimental results indicate that the performance of the proposed method is obviously better than that of previous method, in both color difference error and spectral reflectance error.