The requirements on flexographic spot-color matching model is put forward due to the rapid development of flexography printing and the wide use of environmentally friendly aqueous ink and corrugated paper. In this paper, the flexographic spot-color matching model was designed using BP neural network algorithm for flexography printing where the aqueous ink and corrugated were used. The training and testing samples were obtained by using IGT, and the data was trained based on several mathematics models to find a suitable weighting factor. The matching models' performance and prediction error were analyzed, and the improved algorithm was put forward according to the BP neural network. It showed that the improved BP algorithm was better than the other algorithms in the area of convergence speed and training accuracy.
Xiaozhou Li, Jingqiang Jia, Mingming Cui, Yu Liu, "Research on Several Models of Computer Color Matching for Flexographic Printing Based on Improved BP Neural Network" in Proc. IS&T Printing for Fabrication: Int'l Conf. on Digital Printing Technologies (NIP34), 2018, pp 95 - 98, https://doi.org/10.2352/ISSN.2169-4451.2018.34.95