How can I find the 'average' of two columns of colors in R?

问题

I have a data frame of colors and I would like to find a way to create a third column. The third column should be comprised of a color that is a blend of the two other columns. I have an example data frame and a function that I thought would do the trick. It uses R's colorRampPalette() function and purrr::map2(). However, this returns a listed column that when unlisted is identical to the first column in the dataframe.

# Create a function for colors

x <- RColorBrewer::brewer.pal(11, 'Spectral')

spectral_col <- colorRampPalette(x)

# Apply this function to 

tibble(first = sample(spectral_col(100)),
       second = sample(spectral_col(100))) %>% 
  mutate(middle.color = map2(first, second, function(x, y){
    k <- colorRampPalette(c(x, y))
    k(1)

  }))

Any ideas?

回答1:

If you just generate a palette of length 1, it's always going to be the first value. As in:

colorRampPalette(c("#000000","#FFFFFF"))(1)
#[1] "#000000"

You need to generate at least 3 values to take the middle one between the two specified colours:

colorRampPalette(c("#000000","#FFFFFF"))(3)
#[1] "#000000" "#7F7F7F" "#FFFFFF"
colorRampPalette(c("#000000","#FFFFFF"))(3)[2]
#[1] "#7F7F7F"

You could also use colorRamp as an alternative if you want to specify more precise mid-points:

rgb(colorRamp(c("#000000","#FFFFFF"))(0.5), max=255)
#[1] "#7F7F7F"

With a hat-tip to r2evans, you can put this in your purrr function:

dat <- tibble(first = sample(spectral_col(100)),
       second = sample(spectral_col(100)))

dat %>%
  mutate(middle=map2_chr(first,second, ~rgb(colorRamp(c(.x,.y))(0.5), max=255)))

Or in base R's mapply:

mapply(function(x,y) rgb(colorRamp(c(x,y))(0.5), max=255), dat$first, dat$second)

来源：https://stackoverflow.com/questions/51886163/how-can-i-find-the-average-of-two-columns-of-colors-in-r

标签

function

colors

purrr