杨涛,廖宁放,吴威敏
Yang Tao, Liao Ningfang and Wu WeiminNational Lab of Color Science and Engineering, Beijing Institute of Technology, Beijing 100081, China

• 1:北京理工大学颜色科学与工程国家专业实验室
• 2:北京理工大学颜色科学与工程国家专业实验室
• 3:北京理工大学颜色科学与工程国家专业实验室 北京100083
• 4:北京100083

摘要(Abstract)：

本文提出了一种基于多层BP人工神经网络的水性涂料电脑配色方法。实验表明,BP神经网络模型经训练后能够较好地反映水性涂料反射样本色度参数空间到配方参数空间的非线性映射关系。
A method for paint color matching which is based on back-propagation (BP) neural networks is presented in this paper. The theory of the multi-layer BP neural networks model for the color recipe mapping is introduced, and the experiments for cold water paints are carried out. In the experiments, a set of color chips in the KEYTEC 1190-TC color atlas was used to discuss the feasibility of the model. In addition, BP neural networks were trained to help the realization of the mapping between CIEXYZ space and recipes space for data generalization. The current results show that the networks performed well in the recipes mapping.

关键词(KeyWords)： 电脑配色;水性涂料;BP神经网络
computer color matching, paints color, BP neural networks

Abstract:

Keywords:

基金项目(Foundation):

作者(Author): 杨涛,廖宁放,吴威敏
Yang Tao, Liao Ningfang and Wu WeiminNational Lab of Color Science and Engineering, Beijing Institute of Technology, Beijing 100081, China

参考文献(References)：

• [1]　StephenWestland,LauraIovine,JohnM Bishop:Kubelka MunkorNeuralNetworksfor ComputerColorantFormulation.SPIEVol.4421,745748(2002).
• [2]　LiaoNingfang,YangWeipingetal.CRTColor ConversionbyaMulti LayerBPNeuralNetworks.JournalofImageandGraphics,Vol.5(A)No.6(2000).
• [3]　LiaoNingfang,DouLiumingetal.Computercolormatchingofpowderpaintsusingneural networks.OpticalTechnique,Vol.31No.1(2005).
• [4]　ZengHua,LiaoNingfangetal.Calculation modelusedforreciperevisionincomputercolor matching.SocietyofPhoto OpticalInstrumenta tionEngineers,19961999(1999).
• [5]　HuHinbin,TangXuqing:BPalgorithmandits applicationinartificialneuralnetwork.InformationTechnology,Vol.28No.4,14(2004).