MakeColors: Piecewise color mapping and group-associated coloring

In benja0x40/Barbouille: Visualisation of empirical distributions

Piecewise color mapping and group-associated coloring

Description

MakeColors builds a vector of colors that represent numerical values or group memberships or a combination of both (see the Details section).

Usage

MakeColors(
  v,
  thresholds = NULL,
  colors = NULL,
  range = NULL,
  number = NULL,
  below = "white",
  above = "white",
  na = NA,
  levels = 256,
  name = "",
  parameters = NULL,
  alpha = 1,
  override = TRUE,
  grp = NULL,
  grp.prm = NULL
)

Arguments

v	vector of numeric values.
thresholds	numeric vector defining the boundaries for each range of values to be mapped into a specific color interval. When unspecified, thresholds are automatically set according to the range and the colors or number parameters.
colors	color vector providing the colors representing each threshold value (default = rainbow).
range	range of the numeric values to be represented (default = range(thresholds)).
number	the number of color intervals, used only if thresholds and colors are unspecified (default = 2 in that case).
below	color used for all values below the minimum threshold (defaut = NA, not visible).
above	color used for all values above the maximum threshold (defaut = NA, not visible).
na	color used for missing (NA) values (defaut = NA, not visible).
levels	integer vector controlling the number of color levels generated in each color interval (default = 256, minimum = 2).
name	recall a predefined set of color mapping parameters.
parameters	list of color mapping parameters defined by DefineColorMap.
alpha	transparency (numeric value(s) between 0 and 1).
override	logical determining if the transparency of below and above colors should override the provided alpha value(s) (default = TRUE, yes).
grp	group memberships (default = none). See GroupIndex for documentation on how to specify group memberships.
grp.prm	data.frame of group representation parameters defined by DefineGroupStyles.

Details

The color representation of numerical values is generated by applying piecewise color mapping to these values, with precise control of the underlying color scale. Group coloring consist in using specific colors to represent predefined group memberships. See DefineGroupStyles for the documentation on group representation parameters and GroupIndex for documentation on how to specify group memberships in a finite set of elements. Group coloring can be used exclusively by leaving all the piecewise color mapping parameters unspecified (which implies not providing any values for the first argument of this function).

Value

MakeColors returns a character vector of RGBA colors in hexadecimal.

See Also

ColorLegend, DefineColorMap, DefineGroupStyles, colorize

Examples

# Radial color gradients with two normally distributed random variables
layout(matrix(1:4, 2, 2, byrow = TRUE))

x <- rnorm(2000)
y <- rnorm(2000)
z <- sqrt(x^2 + y^2)
a <- atan2(y, x) + pi / 2

alpha <- ifelse(cos(8 * a) > 0, 0, 1)

clr.prm <- DefineColorMap(
  c(0, 2), c("black", "white"), above = "red", range = c(0, 2.3)
)
clr <- MakeColors(z, parameters = clr.prm, alpha = alpha, override = FALSE)
plot(x, y, xlim = c(-5, 5), ylim = c(-5, 5), pch = 20, col = clr)
ColorLegend("topleft", parameters = clr.prm, cex = 0.8)

clr.prm <- DefineColorMap(
  seq(0, 2, 0.5), c("black", "lightgrey"), above = "red", range = c(0, 2.3)
)
clr <- MakeColors(z, parameters = clr.prm)
plot(x, y, xlim = c(-5, 5), ylim = c(-5, 5), pch = 20, col = clr)
ColorLegend("topleft", parameters = clr.prm, cex = 0.8)

clr.prm <- DefineColorMap(
  thresholds = c(0, 2), colors = c("black", "lightgrey"), above = "red",
  range = c(0, 2.3), levels = 4
)
clr <- MakeColors(z, parameters = clr.prm)
plot(x, y, xlim = c(-5, 5), ylim = c(-5, 5), pch = 20, col = clr)
ColorLegend(
  "topleft", parameters = clr.prm,
  ticks = seq(0, 2, length.out = 5),
  cex = 0.8
)

# Use of a log scale in the color space
x <- runif(2000, -3, 3)
y <- runif(2000, -3, 3)
z <- 1 / sqrt(x^2 + y^2)                    # color mapped values
q <- 1/c(2, 1, 0.5)                         # thresholds
r <- exp(log(1/c(2, 0.5)) + c(-0.25, 0.25)) # range
clr.prm <- DefineColorMap(
  q, c(grey(1:0/2), "red"), above = "lightgrey", below = "lightgrey",
  range = r
)
clr <- MakeColors(z, parameters = clr.prm)
plot(x, y, xlim = c(-5, 5), ylim = c(-5, 5), pch = 20, col = clr)
ColorLegend("topleft", parameters = clr.prm, cex = 0.8, log = TRUE)

benja0x40/Barbouille documentation built on March 26, 2023, 11:38 p.m.