Developing The Characterization Model Of Color Printing System Using Spectral-based Neugebauer Equations

In the printing industry, color separation has always been a key issue in predicting ink combination for rendering a given color. Identifying printer characteristics is very important for solving this problem. With the development of the spectral imaging technology, reconstructing spectral reflectance of a color target for reproduction offers many advantages, such as higher accuracy and without problems associated with metamerism. Hereby the device characterization of color printing system is the research focus in the color science and imaging technology, especially for the cross-media color reproduction.The cellular Yule-Nielsen spectral Neugebauer (CYNSN) model was used to characterize a typical CMYK (i.e., cyan, magenta, yellow, and black) 4-ink color printer. Firstly, the printer to be characterized was investigated in terms of stability, uniformity, tone reproduction, and repeatability. Secondly, the forward CYNSN model was implemented and its performances with different parameters were evaluated so that the optimized parameters were determined. Meanwhile the spectral based model were verified to be more accurate with comparison to CIE1931 XYZ tristimulus values based model in terms of model accuracy and processing speed. Thirdly, the inverse CYNSN model was established based on linear regression iteration and a novel cell searching algorithm was proposed in this dissertation. Large number of high quality hard copy print samples were produced and measured by a spectrophotometer for the model building and assessment. The testing experiments indicated that both forward and inverse models performed with high spectral and colorimetric accuracy.The phenomenon of ink density additivity failure was discussed and the sub-additivity equations were conducted for correcting it. Based on the spectral density sub-additivity equations, the printer was characterized by the grey component replacement (GCR) model and the accurate spectral match was achieved.Finally, the main work of this dissertation was summarized, and the perspective of the future study was also forecasted.