A new adaptive color controller for image scanning and printing processes is proposed. The objective of controlling colors is to achieve the desired accuracy of color reproduction under various viewing and graphic-arts conditions. Based on the plant inverse approach, the adaptive color controller is devised to eliminate color errors for image reproduction. And, it involves a novel feature-extraction method and a model-free estimator. In this approach, the feature-extraction method characterizes the extreme nonlinearity of any combination of color scanning and printing devices in terms of the human vision characteristics. For the consideration of easy implementation and fast image-processing speed, the higher-order CMAC neural network, which was proposed by Lane in 1991, is employed to work for the model-free approximation. In this paper, the colorimetric simulation illustrates the satisfactory performance. Finally, its effectiveness is also demonstrated by several colorimetric experiments.
|頁（從 - 到）||997-1001|
|期刊||Proceedings - International Conference on Image Processing, ICIP|
|出版狀態||已發佈 - 1994 一月 1|
|事件||Proceedings of the 1994 1st IEEE International Conference on Image Processing. Part 3 (of 3) - Austin, TX, USA|
持續時間: 1994 十一月 13 → 1994 十一月 16
ASJC Scopus subject areas