optimize_palette: optimize raw palette obtained from create_palette

View source: R/optimize_palette.R

optimize_paletteR Documentation

optimize raw palette obtained from create_palette

Description

taken a raw palette created from a jpeg image, optimizes it given the type of variable to be plotted

Usage

optimize_palette(
  rgb_raw_palette = NA,
  number_of_colors = NA,
  type_of_variable = "categorical",
  effective_n_of_color = NA,
  filter_on_low_brightness = NA,
  filter_on_high_brightness = NA,
  filter_on_saturation = NA
)

Arguments

rgb_raw_palette

numeric matrix of RGB from a call to rgb on hex codes

type_of_variable

string, default to 'categorical'. type of variable to be plotted with the building palette

effective_n_of_color

integere, the actual number of colors obtained from the application of kmeans on the image. equal to number_of_colours *100

filter_on_low_brightness

boolean, default to true. specifies if a filter on colours with low brigthness should be applied to enhance the palette

filter_on_high_brightness

boolean, default to true. specifies if a filter on colours with high brigthness should be applied to enhance the palette

filter_on_saturation

boolean, default to ture. specifies if a filter on low saturation should be applied.

number_of_colours

integer, number of desired colours in the final palette, as specified by the user when calling create_palette

Details

palette optimization consists into four different steps: - conversion to hsv scale in order to easily elavorate on colour order and properties. - filter on colours with a brightness lower than the first quartile of v distribution - filter on colours with a brightness higher or equal to the Tukey's outlier threshold computed on the overall v distribution - subset of the palette according to the type of variable to be plotted: a spaced sample in case of categorical variables, an interpolation between two colours close to the mode of h for continuous variables

Author(s)

Andrea Cirillo

Examples

create_palette("data/nascita_venere.jpg",number_of_colors = 20)

AndreaCirilloAC/paletter documentation built on Jan. 11, 2023, 4:45 a.m.