基于模糊辨识的设备彩色特征化模型的研究

A Device Color Characterization Model Based on Fuzzy Identification

基于颜色自然语言描述的模糊性,以RGB颜色空间与CIEL*a*b*颜色空间转换为例,建立了以模糊控制方法为基础的彩色设备特征化模型,并对照基于三维查找表与插值算法的彩色设备特征化方法,对该模型进行了讨论。研究结果显示,该模型设计简单,便于应用,可以适用于设备彩色的特征化非线性描述。当采样空间点数量为125个时,模型对于检验样本点转换的平均色差达到2.65,而要达到相同水平的平均色差值,三维查找表与插值算法至少需要512个采样空间量,即该模型具有在较少样本点情况下可以获得较高精度的优点。

The device color characteristic the fuzzy properties of natural langua~ description is one of key technologies of color management. Considering ~e description of color, this paper transforms RGB color space to CIEL ＊ a ＊ b ＊ color space, for example, builds the device color characterization model by using the fuzzy control theory and discusses the characterization model by comparing it with the characterization method which is based on 3-dimensional look-up table （3D-LUT） and the interpolation algorithm. The results show that the characterization model is simple for design and convenient for application. It is suitable for the nonlinear description of device color characteristics. The average number of color errors of test sample points obtained by the characterization model is 2. 65 when the number of test sample points is 125. To obtain the same number of average color errors, the 3D-LUT ＆ interpolation algorithm needs at least 512 test samples points. That is to say, compared with the 3D-LUT ＆ inter- polation algorithm, our color characterization model has better precision even when there are few sample points.