r - 将自定义图像添加到 ggplot 构面

Question

我想向每个方面添加自定义图像。使用 annotation_custom 可以在所有方面复制图像，例如:

require(ggplot2); require(grid); require(png); require(RCurl)

p = ggplot(iris, aes(Sepal.Length, Sepal.Width)) + geom_point() + facet_wrap(~Species)

img1 = readPNG(getURLContent('https://cdn2.iconfinder.com/data/icons/animals/48/Turtle.png'))
img2 = readPNG(getURLContent('https://cdn2.iconfinder.com/data/icons/animals/48/Elephant.png'))
img3 = readPNG(getURLContent('https://cdn2.iconfinder.com/data/icons/animals/48/Hippopotamus.png'))

a1 = annotation_custom(rasterGrob(img1, interpolate=TRUE), xmin=7, xmax=8, ymin=3.75, ymax=4.5)
a2 = annotation_custom(rasterGrob(img2, interpolate=TRUE), xmin=7, xmax=8, ymin=3.75, ymax=4.5)
a3 = annotation_custom(rasterGrob(img3, interpolate=TRUE), xmin=7, xmax=8, ymin=3.75, ymax=4.5)

p + a1

是否有替代方法可以实现此目的，以便 img1-3 正确填充相应的方面？

Answer 1

为了完整起见，我添加了答案。所有功劳都归功于@baptiste，他建议使用annotation_custom2函数。

require(ggplot2); require(grid); require(png); require(RCurl)

p = ggplot(iris, aes(Sepal.Length, Sepal.Width)) + geom_point() + facet_wrap(~Species)

img1 = readPNG(getURLContent('https://cdn2.iconfinder.com/data/icons/animals/48/Turtle.png'))
img2 = readPNG(getURLContent('https://cdn2.iconfinder.com/data/icons/animals/48/Elephant.png'))
img3 = readPNG(getURLContent('https://cdn2.iconfinder.com/data/icons/animals/48/Hippopotamus.png'))


annotation_custom2 <- 
function (grob, xmin = -Inf, xmax = Inf, ymin = -Inf, ymax = Inf, data){ layer(data = data, stat = StatIdentity, position = PositionIdentity, 
        geom = ggplot2:::GeomCustomAnn,
        inherit.aes = TRUE, params = list(grob = grob, 
                                          xmin = xmin, xmax = xmax, 
                                          ymin = ymin, ymax = ymax))}

a1 = annotation_custom2(rasterGrob(img1, interpolate=TRUE), xmin=7, xmax=8, ymin=3.75, ymax=4.5, data=iris[1,])
a2 = annotation_custom2(rasterGrob(img2, interpolate=TRUE), xmin=7, xmax=8, ymin=3.75, ymax=4.5, data=iris[51,])
a3 = annotation_custom2(rasterGrob(img3, interpolate=TRUE), xmin=7, xmax=8, ymin=3.75, ymax=4.5, data=iris[101,])

p + a1 + a2 + a3

输出: