Skip to contents

Photoreceptor sensitivity matrix for colourvision

Source: R/colourvision_interop.R
species_sensitivity_matrix.Rd

Assembles a species' validated chromatic receptors into the sensitivity matrix C that colourvision models consume (a data frame whose first column is wavelength and whose remaining columns are one photoreceptor each). The receptors and their Govardovskii templates are exactly those colour_jnd uses, so a colourvision model built on this matrix reproduces luxR's own quantum catches.

Usage

species_sensitivity_matrix(
  species,
  lambda = seq(300, 700, 1),
  receptor = NULL,
  weight = c("photon", "energy")
)

Arguments

species

Character scalar naming a species in species_sensitivities with at least one validated chromatic receptor.

lambda

Numeric wavelength grid (nm); must be finite, strictly increasing, and unique. Defaults to seq(300, 700, 1), colourvision's default nm.

receptor

Optional character vector selecting a subset of the species' validated chromatic receptors; passed through to the same channel resolution colour_jnd uses.

weight

One of "photon" (default, columns \(S\lambda\)) or "energy" (columns \(S\)).

Value

A data.frame with a wl column plus one column per receptor, named and ordered as luxR's channel definitions.

Details

luxR integrates photon catch (proportional to \(E\lambda\)) while colourvision integrates energy (\(\int I R C\)). With weight = "photon" (the default) the columns are the sensitivities multiplied by wavelength (\(S\lambda\)), so colourvision's energy integral reproduces luxR's photon catch; the proportionality constant cancels in the log RNL model. Use weight = "energy" for the raw sensitivities.

See also

as_colourvision, colour_jnd, colourvision::RNLmodel

Examples

if (FALSE) { # \dontrun{
  C <- species_sensitivity_matrix("Apis mellifera")
} # }