ggquadrilateral

ggquadrilateral provides a ggplot2 geom that can draw arbitrary quadrilaterals in a convenient way.

Installation

You can install the latest version of ggquadrilateral from GitHub with:

devtools::install_github("const-ae/ggquadrilateral")

If you don’t already have devtools installed, you can get it from CRAN with install.packages("devtools").

Example

First load ggplot2 and the ggquadrilateral package

library(ggplot2)
library(ggquadrilateral)

The simplest example is to just define the positions of the four corners manually

kite_df <- data.frame(
  left_tip_x = 2,
  left_tip_y = 7,
  top_tip_x = 3,
  top_tip_y = 8,
  right_tip_x = 4,
  right_tip_y = 7,
  bottom_tip_x = 3,
  bottom_tip_y = 3
)
kite_df
#>   left_tip_x left_tip_y top_tip_x top_tip_y right_tip_x right_tip_y
#> 1          2          7         3         8           4           7
#>   bottom_tip_x bottom_tip_y
#> 1            3            3

ggplot(kite_df) +
  geom_quadrilateral(aes(x1=left_tip_x, y1 = left_tip_y,
                         x2 = top_tip_x, y2 = top_tip_y,
                         x3 = right_tip_x, y3 = right_tip_y,
                         x4 = bottom_tip_x, y4 = bottom_tip_y),
                     color = "black", fill = "purple", size=4) +
  xlim(-2, 8) + ylim(0, 10)

It can also be used to visualize more complex data. We will now use it to draw the triangle mesh for 10 random points.

df <- data.frame(x=rnorm(n=10, mean=0, sd=1),
                 y=rnorm(n=10, mean=0, sd=1))

ggplot(df, aes(x=x, y=y)) +
  geom_point()

We will use the tripack package to calculate the Delauney triangulation.

library(tripack)
triang <- as.data.frame(triangles(tri.mesh(df)))
triang_df <- data.frame(id = seq_len(nrow(triang)),
           p1x = df$x[triang$node1],
           p1y = df$y[triang$node1],
           p2x = df$x[triang$node2],
           p2y = df$y[triang$node2],
           p3x = df$x[triang$node3],
           p3y = df$y[triang$node3])
head(triang_df)
#>   id        p1x       p1y        p2x         p2y        p3x         p3y
#> 1  1 -0.6264538 1.5117812 -0.8204684 -0.04493361 -0.3053884  0.59390132
#> 2  2 -0.6264538 1.5117812 -0.3053884  0.59390132  0.3295078  1.12493092
#> 3  3  0.1836433 0.3898432  0.4874291 -0.01619026  0.5757814  0.82122120
#> 4  4  0.1836433 0.3898432  0.5757814  0.82122120  0.3295078  1.12493092
#> 5  5  0.1836433 0.3898432  0.3295078  1.12493092 -0.3053884  0.59390132
#> 6  6  0.1836433 0.3898432 -0.3053884  0.59390132 -0.8204684 -0.04493361

Using the triang_df that the coordinates for the 12 interpolating triangles we can make the plot:

ggplot() +
  geom_quadrilateral(data=triang_df, 
                     mapping = aes(
                       x1 = p1x, y1 = p1y,
                       x2 = p2x, y2 = p2y,
                       x3 = p3x, y3 = p3y,
                       # We want triangles, so we make the
                       # fourth point identical to the third
                       x4 = p3x, y4 = p3y,
                       fill = as.factor(id)),
                     color = "black") +
  geom_point(data= df, aes(x=x, y=y), size = 3)