<title>Gamut mapping in multiple color spaces</title>

Abstract

In the process of gamut mapping from monitor display into printer hardcopy in CIE L*a*b* color space, blue is tend to map to purple. This paper presents a new approach to solve the perceived blue hue shift problem. By this approach, the entire color gamut is divided into four regions: a non-blue region, a blue-region, and two in-between regions. The segmentation of the four regions is based on the hue angle in CIE L*a*b* color space. Different color spaces are applied to different regions for gamut mapping. In the non-blue region, CIE L*a*b* color space is applied for gamut mapping. In the blue region, CIE L*u*v* color space is applied to eliminate the perceived blue shift. In the two in-between regions, both color spaces are used for gamut mapping, and a weighting function is applied for smooth transaction.Three mapping techniques to test this mapping approach are: lightness rescaling followed by croma compression in constant lightness direction, lightness rescaling followed by croma compression along the straight line connected by the current color and the mid-point of the achromatic line, and lightness rescaling followed by minimum distance mapping. The experimental results show that the blue shift disappears in the blue region, and the color mapping is not changed in the non-blue region.This approach can be extended to applying more than two color spaces for gamut mapping. It is also suitable for optimal gamut mapping and preference gamut mapping.

Spring-primary mapping: a fast color mapping method for primary adjustment and gamut mapping

“…In our implementation, we set following default parameters, (w L , w h ) for the six primaries to determine aimed primaries: R (0.5, 0.5), G(0.75, 0.5), B(0.75, 0), C(0.75, 0.5), M(0.75, 0.75), Y (1,1). They are applied to generate sRGB color maps for the tradeoff between pictorials and business graphics.…”

“…Without hue rotation, blue colors are mapped to purple during the gamut mapping from a monitor gamut to a printer gamut [1]. It was found that this problem came from the serious blue hue nonlinearity in CIELAB color space [2].…”

“…To solve color space problems (e.g. hue-linearity and color uniformity) for gamut mapping, Zeng proposed using multiple color spaces for gamut mapping [1]. For example, a blue-hue linearized color space is applied for gamut mapping in the blue region, CIELAB color space is applied for gamut mapping for other regions, and the combination of both color spaces is applied for the gamut mapping in transition regions for smooth transitions.…”

A colour printer calibration method based on gamut division algorithms

“…Thus, it must store a large amount of sample data, which limits computational efficiency. Zeng has developed a new GCR algorithm by directly converting CIEXYZ values to CMYK values. In this algorithm, the K amount can be flexibly controlled by users, but the precision is highly influenced by the additivity of colorimetric density.…”