This example calculates the XYZ tristimulus values of the D65 illuminant for both the CIE 1931 2º standard observer and the CIE 2015 10º observer.

  use colorimetry::illuminant::D65;

  // D65 Tristimulus values, using the CIE1931 standard observer by default
  let xyz_d65 = D65.xyz(None).set_illuminance(100.0);

  let [x, y, z] = xyz_d65.values();
  // [95.04, 100.0, 108.86]

  // D65 Tristimulus values using the CIE2015 10º observer
  use colorimetry::observer::Observer::Cie2015_10;
  let xyz_d65_10 = D65
    .xyz(Some(Cie2015_10)).set_illuminance(100.0);

  let [x_10, y_10, z_10] = xyz_d65_10.values();
  //[94.72, 100.0, 107.143]

The correlated color temperature (CCT) of an illuminant, typically expressed in kelvin (K), describes whether a light source appears warm (low CCT) or cool (high CCT). It is a key parameter for characterizing the visual appearance of white light . This example calculates both the correlated color temperature and the deviation from the Planckian locus, often referred to as the tint.

  use colorimetry::illuminant::A;

  // Calculate CCT and Duv for the A illuminant
  // Requires `cct`, and `cie-illuminants` features
  let [cct, duv] = A.cct().unwrap().values();
  // [2855.4977, 0.0]

The CIE has announced that the Color Fidelity Index (CFI) will replace the Color Rendering Index (CRI) as the standard metric for evaluating color rendering. Both indices aim to quantify how accurately a light source reproduces the colors of illuminated objects. However, the CFI offers a significant improvement in accuracy by using 99 reference color samples and more advanced color difference metrics, compared to the CRI’s use of only 8 samples. Below is an example calculation of the general Color Fidelity Index for the CIE F2 illuminant:

  use colorimetry::illuminant::F2;

  // Calculate the Color Fidelity Index of the CIE F2 standard illuminant
  // Requires `cfi`, and `cie-illuminants` features
  let cf_f2 = F2.cfi().unwrap();
  let cf = cf_f2.general_color_fidelity_index();
  // 70.3

The spectral locus is the boundary in a chromaticity diagram that encloses all perceivable, physically realizable colors. Due to its shape, it is sometimes informally referred to as the "horseshoe." Below, we compute the chromaticity coordinates that define the spectral locus.

  use colorimetry::observer::Observer::Cie1931;
  let mut locus = Vec::new();
  let wavelength_range = Cie1931.spectral_locus_wavelength_range();
  for wavelength in wavelength_range {
    // unwrap OK because nm is in range
    let xyz = Cie1931.xyz_at_wavelength(wavelength).unwrap();
    let chromaticity = xyz.chromaticity();
    locus.push([wavelength as f64, chromaticity.x(), chromaticity.y()]);
  }
  println!("{locus:?}");

This is usually done with the CIE 1931 Standard Observer, but this library supports any observer—as long as both the color space and the data use the same one. Instead of fixed XYZ values, it computes conversions from the spectral definitions of the primaries to be able to do so. Here, we compute transformation matrices for the DisplayP3 color space using both the Cie1931 and Cie2015 observers.

  use colorimetry::observer::Observer;
  use colorimetry::rgb::RgbSpace::DisplayP3;

  let xyz2rgb_31 = Observer::Cie1931.xyz2rgb_matrix(DisplayP3);
  //  2.4933, -0.9313, -0.4027,
  // -0.8298,  1.7629,  0.0236,
  //  0.0355, -0.076,   0.9574

  let rgb2xyz_31 = Observer::Cie1931.rgb2xyz_matrix(DisplayP3);
  // 0.4866, 0.2656, 0.1981,
  // 0.2291, 0.6917, 0.0792,
  // 0.0001, 0.0451, 1.0433,

  use colorimetry::observer::Observer::Cie2015;

  let xyz2rgb_15 = Cie2015.xyz2rgb_matrix(DisplayP3).clone();
  //  2.5258,  -1.0009, -0.3649,
  // -0.9006,   1.8546, -0.0011,
  //  0.0279,  -0.0574,  0.95874

This example finds the best matching color in the Munsell Color Book for a given sample—in this case, the R9 test color used in the CRI color rendering method. It uses the CieCam16::de_ucs color difference metric and the Cie2015_10 standard observer to calculate perceptual similarity.

The closest match identified is Munsell "5R 5/14", a vivid red hue, with a color difference of just 3 ΔE. In practical terms, a ΔE of 3 is considered a close match—just at the threshold where most observers might start to notice a difference under controlled viewing conditions.

  // requires `cri` and `munsell` features
  use colorimetry::observer::Observer::Cie2015_10;
  use colorimetry::colorant::{MunsellCollection, TCS};

  let cri_r9 = &TCS[8];
  let (key, delta_e) = MunsellCollection::match_ciecam16(
    cri_r9,
    None,
    None,
    Some(Cie2015_10),
  ).unwrap();
  // ("5R4/14", 2.85)

This example matches the Munsell paint chip 5 BG 5/8—a teal/blue-green color— to its nearest sRGB rgb(0, 113, 138) equivalent, mimicking real-world viewing conditions.

Instead of the traditional CIE 1931 2° observer, this match uses the CIE 2015 10° observer, which more accurately reflects how paint colors appear on walls. The illumination is based on the warm-white LED_B2 standard illuminant (≈ 3000 K). Together, these adjustments help the display color reflect what you'd actually see on a freshly painted surface.

  // requires `cie-illuminants`, and `munsell` features
  use colorimetry::{
    cam::{ViewConditions, CIE248_HOME_SCREEN},
    colorant::Munsell,
    illuminant::LED_B2,
    observer::Observer::{Cie1931, Cie2015_10},
    rgb::RgbSpace::SRGB,
  };

  let paint = Munsell::try_new("5BG5/8").unwrap();
  let vc = ViewConditions::average_surround(6.0);
  let cam_paint = Cie2015_10.ciecam16(&LED_B2, &paint, vc);
  let rgb_2015 = cam_paint
    .rgb(SRGB, Some(CIE248_HOME_SCREEN))
    .unwrap()
    .compress();

  // Use a spectral representation of the Cie2015_10 RGB pixel, using the `Rgb`'s Light trait,
  // and calculate its XYZ tristimulus and RGB values for the CIE 1931 standard observer, the
  // observer
  // required for the sRGB color space.
  let xyz_1931 = Cie1931.xyz(&rgb_2015, None);
  let rgb_1931 = xyz_1931.rgb(SRGB).compress();
  let [r, g, b]: [u8; 3] = rgb_1931.into();
  //  (0, 113, 138)

To enable a feature, such as cri and munsell, use

cargo add colorimetry -F cri,munsell

or, if you prefer to use the cargo add command with the --features flag, you can run:

cargo add colorimetry --features cri,munsell

Alternatively, configure features manually in your Cargo.toml:

colorimetry = { version = "0.0.7", features = ["cri", "munsell"] }