Color Science - ISET/isetcam GitHub Wiki

• Supporting Concepts overview
• Image processing, color, metrics, and displays tutorials
• Metrics

---

Color is computed throughout the pipeline, not one stage of it

Color science does not live in a single ISETCam structure. It connects scene illuminants, sensor color filters, image-processing color balance, display primaries, and image-quality metrics — see the Core Pipeline pages for each of those. This page collects the shared colorimetric building blocks those pages all rely on.

Illuminants

Scene illumination (see Scene Radiance) is created with:

bb  = blackbody(400:10:700, 5500, 'energy');   % 5500 K blackbody radiator
ill = illuminantCreate('d65', 400:10:700);     % Named illuminant, e.g. daylight
d   = daylight(400:10:700, 6500);              % CIE daylight series at a given CCT

t_colorEnergyQuanta (see Tutorials) covers converting between energy and photon-quanta units for these spectra.

Color matching and chromaticity

ieXYZFromPhotons converts spectral radiance or irradiance (in photons) into CIE XYZ tristimulus values, using the CIE color matching functions. From XYZ, xyy2xyz and xyz2uv compute the chromaticity coordinates plotted in a standard CIE chromaticity diagram. t_colorMatching and t_colorChromaticity (see Tutorials) work through both.

Colorimetric transforms

color/transforms/ implements the standard conversions between the color spaces ISETCam uses at different pipeline stages:

From -> To	Function
XYZ <-> CIELAB	ieXYZ2LAB, ieLAB2XYZ
XYZ <-> LMS (cone)	xyz2lms, lms2xyz
XYZ <-> sRGB	xyz2srgb, srgb2xyz
linear RGB <-> sRGB	lrgb2srgb, srgb2lrgb
LMS -> sRGB	lms2srgb
xyY <-> XYZ	xyy2xyz
sRGB -> correlated color temperature	srgb2colortemp

colorTransformMatrix / colorTransformMatrixCreate build the linear transforms (for example, sensor-to-XYZ) used by the image processor's sensor conversion methods.

Recommended tutorials and examples

• Image processing, color, metrics, and displays — t_colorMatching, t_colorChromaticity, and t_colorEnergyQuanta.
• Examples — examples/color/ has further scripts, including s_srgbGamut (display gamut) and s_colorIlluminantTransforms (blackbody illuminant correction).

Related FISE concepts

The online Foundations of Image Systems Engineering (FISE) develops color matching, chromaticity, and color spaces in more depth as conceptual background.

Boundaries: human color vision

The functions above describe colorimetry — how spectra map to standard color spaces such as XYZ and CIELAB — not the biological mechanisms of human color vision (cone spectral sensitivities, opponent processes, color appearance). For detailed modeling of the human visual system, see ISETBio.