CMAC辨识的CMYK到CIE L*a*b*颜色空间转换模型的研究

为了解决设备相关颜色空间CMYK与设备无关颜色空间之间的相互转换问题,利用小脑模型神经网络(cerebellar model articulation controller, CMAC)高度非线性拟合能力,研究CMYK颜色空间与CIE L*a*b*之间的转换关系,研究结果显示该方法具有结构简单,易于软件和硬件的实现,将IT8.7/3标准色靶文件中104个专业色块值作为检验样本,检验样本的平均色差为1.6,完全适用于两种不同颜色空间之间的转换过程. 

Research on Color Space Conversion Model of CMYK to CIE L* a* b* Based on CMAC Identification

For solveing the conversional issuse on device-dependend color space CMYK and the device-independent color space, the highly nonlinear fitting ability on cerebellar model articulation controller (CMAC) neural network was used to research the conversion relationship between CMYK and CIE L^*a^*b^* color space in this paper. The results show that the method has a simple structruc, and can be easily implemented by hardware and software. The average color error of the test samples is 1.6 when taking the 104 professional color values of the IT8.7/3 standard color target file as the test sample, which further shows that the proposed method can be applied to the conversion prccess between the two different color spaces.