Voronoise

The goal of voronoise is to make pretty pictures. It is a very minor tweak on the voronoi tesselations provided by the ggforce package. You can install the development version of voronoise from GitHub with:

remotes::install_github("djnavarro/voronoise")

Example

Create a tibble with columns specifying the x and y coordinates, and the colours to be associated with the corresponding Voronoi cell

library(voronoise)
#> Loading required package: ggplot2
#> Loading required package: ggforce

set.seed(1)

dat <- tibble::tibble(
  x = runif(n = 50, min = .1, max = .9),
  y = runif(n = 50, min = .1, max = .9),
  shade = sample(colours(), size = 50, replace = TRUE)
)

dat
#> # A tibble: 50 x 3
#>        x     y shade       
#>    <dbl> <dbl> <chr>       
#>  1 0.312 0.482 tan         
#>  2 0.398 0.789 grey43      
#>  3 0.558 0.450 plum4       
#>  4 0.827 0.296 rosybrown   
#>  5 0.261 0.157 grey26      
#>  6 0.819 0.180 springgreen4
#>  7 0.856 0.353 gold3       
#>  8 0.629 0.515 grey68      
#>  9 0.603 0.630 navajowhite2
#> 10 0.149 0.425 tomato      
#> # … with 40 more rows

The voronoise_base() function initialises a ggplot object

The default behaviour of geom_voronoise() is identical to ggforce::geom_voronoi_tile() with some parameter values changed. It does not perturb the location of any of the tiles:

voronoise_base(dat) +
  geom_voronoise(fill = "antiquewhite2")

To perturb the tiles we need to pass it a perturb function:

voronoise_base(dat) +
  geom_voronoise(fill = "antiquewhite2") +
  geom_voronoise(perturb = perturb_uniform(.2))

Note that perturb_uniform() returns a function. The voronoise package comes with a couple of handy functions that will create custom perturbing functions. For example:

voronoise_base(dat) +
  geom_voronoise(fill = "antiquewhite2") +
  geom_voronoise(perturb = perturb_float(-90))

An example using some more settings:

set.seed(1)
voronoise_base(
  data = voronoise_data(100, viridis::magma(10)),
  background = "skyblue3"
) +
  geom_voronoise(fill = "skyblue2") +
  geom_voronoise(
    perturb = perturb_float(
      angles = c(0, 90, 180),
      noise = c(2, 1)
    )
  )