Computes, from the hex colors alone, how suitable each palette is for the
common plotting jobs. Used by tf_find_palettes(); exposed so you can
apply your own thresholds.
Usage
tf_palette_stats(names = tf_palette_names(medium = "all"))Value
A data frame with one row per palette and the columns:
- palette
Palette name.
- n_colours
Number of colors.
- min_de
Smallest pairwise Lab distance between any two colors. Larger is easier to tell apart; 20 is a comfortable floor for categorical use.
- min_de_deutan, min_de_protan, min_de_tritan
The same after simulating each color-vision deficiency.
- lum_range
Range of CIE lightness (0 to 100) across the palette.
- hue_spread
Angular spread of the chromatic colors' hues, in degrees. Near 0 means one hue family.
- hue_clusters
Number of distinct hue families (hues separated by more than 60 degrees).
- has_light
Whether the palette contains a light or neutral color that can serve as a midpoint.
- categorical, colorblind, sequential, diverging
Logical flags as used by
tf_find_palettes().
Details
Distances are CIE76 (Euclidean in CIE Lab). Color-vision deficiency is simulated with the Machado, Oliveira and Fernandes (2009) matrices at full severity for deuteranopia, protanopia and tritanopia. Hue statistics only consider colors with chroma of at least 20, so grays and blacks do not count as a hue.
Examples
s <- tf_palette_stats()
head(s[order(-s$min_de_deutan), c("palette", "min_de", "min_de_deutan")])
#> palette min_de min_de_deutan
#> 277 scattershot_comic 60.24956 60.12099
#> 248 air_raid_comic 60.16312 58.55659
#> 249 fireflight_comic 60.16312 58.55659
#> 251 slingshot_comic 60.16312 58.55659
#> 204 mirage_cartoon 57.99994 53.21237
#> 119 ratbat 59.59306 51.21162