Application of Artificial Neural Network in CCD Device Characterization for Press Colors

摘要：

To capture and make assessment of printing colors, industrial CCD device requires transform relationship from RGB response values to CIE colorimetric values. Design and make a color target covering the whole printing gamut according to printing color natures. Based on this color target, a BP neural network model is introduced to establish the relationship from RGB color space to CIE color space. As is justified in this experiment, with one hidden layer, 16 hidden layer cells and tansig transfer function, the neural network is able to obtain a steady relationship with average and maximal color difference of 1.67 and 12.42 respectively, better than that caused by polynomial modeling method under the same condition.

作者: Guo Ge Xu Yanfang

作者单位: School of Printing & Packing Engineering Beijing Institute of Graphic Communication Beijing, P.R.Chi School of Printing & Packing Engineering Beijing Institute of Graphic Communication Beijing, P.R.C

会议类型: 国际会议

会议名称: The 31st International Congress on Imaging Science(第31届国际影像科学大会 ICIS2010)

会议地点: 北京

会议语种:英文

页码: 257-259

在线出版日期: 2010-05-12（万方平台首次上网日期，不代表论文的发表时间）