Colourful supports working with correlated color temperature (CCT) in two ways:
- Approximating the xy chromaticity from a CCT specified in kelvin (K).
- Approximating the CCT from a given xy chromaticity.
Your input in this case is a temperature in kelvin (K), and you want to figure out what color this corresponds to.
Colourful provides a helper class CCTConverter that can be used to approximate the xy chromaticity from the given temperature.
double temperature = 3000; // in K
xyChromaticity chromaticity = CCTConverter.GetChromaticityOfCCT(temperature); // xy [x=0.44, y=0.4]Using this, you'll only have the chromaticity, not an actual color. However, you can use a color converter to convert this to the target color space. In the example below, we'll try to represent the color in the sRGB color space.
IColorConverter<xyChromaticity, RGBColor> converter = new ConverterBuilder().Fromxy(Illuminants.D65).ToRGB(RGBWorkingSpaces.sRGB).Build();
RGBColor color = converter.Convert(chromaticity); // RGB [R=1.28, G=0.93, B=0.56]You'll notice that the resulting RGB color is outside of its range between 0 and 1, the red channel exceeds this. This is because the input chromaticity doesn't have a luminance associated with it, and during the color conversion, it used the maximum luminance of 1 in the xyY color space. And the resulting color is outside of the range. To get a value inside the RGB range, we can divide the individual channel by the highest channel value (in this case it's 1.28). This will make at least one of the channels with the maximum value 1, and still keep the right hue. In Colourful, there's a helper for this for RGB colors called NormalizeIntensity.
color = color.NormalizeIntensity(); // RGB [R=1, G=0.72, B=0.44]This value corresponds to RGB values 255, 185, 111, or #FFB96F in the hex format.
To approximate chromaticity from CCT, Colourful currently uses the Planckian locus approximation method. The algorithm is described here.
Alternative methods aren't yet implemented (see #74).
Your input in this case is a color and you want to approximate what correlated color temperature (CCT) in kelvin (K) it corresponds to.
RGBColor color = RGBColor.FromRGB8Bit(255, 121, 0);
IColorConverter<RGBColor, xyChromaticity> converter = new ConverterBuilder().FromRGB(RGBWorkingSpaces.sRGB).Toxy(Illuminants.D65).Build();
xyChromaticity chromaticity = converter.Convert(color); // xy [x=0.55, y=0.4]
double temperature = CCTConverter.GetCCTOfChromaticity(chromaticity); // 1293 KThe results are unfortunately not too precise and sometimes you might end up with values outside of the expected range.
To approximate CCT from chromaticity, Colourful currently uses low-temperature equation proposed by J. Hern谩ndez-Andr茅s, R. L. Lee, and J. Romero. The algorithm is described here.
Alternative methods aren't yet implemented (see #74).