Device- and illuminant-independent color reproduction using principal component analysis and neural networks

摘要

There exist several attempts to develop a system for device-independent color reproduction in desktop publishing. However, because conventional color management system is calibrated only by the colorimetric values under the standard illuminant, perceptual difference between the original and the reproduced colors will occur when these colors are viewed under non-standard illuminant. In this study, we present a method for device-and-illuminant independent color reproduction by adopting the spectral reflectance as an intermediate color representation. We trained three layered neural networks to realize the transformation from CMY values of the proof printer to the principal components (PCs) of the spectral reflectance, and the transformation from these PCs to C$PRM@M$PRM@Y$PRM values of dye-sublimation printer. after the learning process, we evaluated the accuracy of the transformation from CMY to C$PRM@M$PRM@Y$PRM through PCs by the trained neural networks.