Abstract
The color rendition problem was discussed from a standpoint of image processing system. In conventional color printing, Neugebauer equation is known as a theoretical model for color reproduction. This model is based on colored-area analysis of halftone dots and resulted in good prediction for colors reproduced by binary dots such as thermal ink-transfer printing. However, it was not always useful for more complicated dot formation mechanism such as ink-jet printing. In this paper, a black-box color matching model is introduced for practical applications. The color correction functions, expanded by linear and non-linear polinomials, were estimated in actual examples. It is concluded that second order color correction, where the matching coefficients are optimized by the method of least squares, should be performed for better color rendition. Two examples of pictorial color hard-imaging systems are presented. The one is equipped with a special purpose color signal processor for realtime execution of color-tone correction. The other is composed of general purpose image processor and micro-computer. A novel color correction technique, where the cross terms of correction functions are expressed by additive square formula, could be applied for economical computation of second order color masking equation.
