Script that can visualize how images (like banners or sprays) would look in Minecraft 1.8

BannerDisplay.jl

using Colors
using Images
using ProgressMeter
using Distributed

# 1.8 base colors, see https://minecraft.gamepedia.com/Map_item_format#Map_colors
baseColors = [0 0 0;
    127 178 56;
    247 233 163;
    167 167 167;
    255 0 0;
    160 160 255;
    167 167 167;
    0 124 0;
    255 255 255;
    164 168 184;
    183 106 47;
    112 112 112;
    64 64 255;
    104 83 50;
    255 252 245;
    216 127 51;
    178 76 216;
    102 153 216;
    229 229 51;
    127 204 25;
    242 127 165;
    76 76 76;
    153 153 153;
    76 127 153;
    127 63 178;
    51 76 178;
    102 76 51;
    102 127 51;
    153 51 51;
    25 25 25;
    250 238 77;
    92 219 213;
    74 128 255;
    0 217 58;
    21 20 31;
    112 2 0;
    126 84 48]

# compute the variants
mapColors = map(eachrow(baseColors)) do color    
    [RGB((floor.(color .* mult) ./ 255)...) for mult in [255 180 220 135] ./ 255]
end

# restructure so that each element of the array is an RGB element corresponding to a color variant
mapColors = hcat(mapColors...)[1, :]

"""
    closestPixel(pixel::RGB; metric=DE_2000())

Maps the pixel (given in RGB from Colors.jl) to the closest permitted color in Minecraft 1.8 using the Delta-E color metric. Optionally
takes another color metric to use. Returns a color from `mapColors`.
"""
function closestPixel(pixel::RGB; m=DE_2000())
    Δ = colordiff.(pixel, mapColors; metric=m)
    Δ₀ = mapColors[argmin(Δ)]
end

"""
    mcConvert(img; metric=DE_2000())

Converts an image using as many available processes as possible (can do distributed computing with the Distributed library).

# Example
Load an image and convert it with 4 processes
```julia-repl
julia> using Distributed
julia> addprocs(4) # or start julia with 'julia -p 4'
julia> @everywhere include("BannerDisplay.jl") # imports this file in all the processes
julia> img = load("test.png")
julia> R = mcConvert(img)
# then can show images with Plots.jl or save them
```
"""
function mcConvert(img; metric=DE_2000())
    img = RGB.(img)

    # Distributed computing, shows progressbar and number of processes.
    println("Starting with $(nworkers()) processes...")
    R = @showprogress pmap(eachcol(img)) do px  # column-wise operations are faster in Julia
        rgb = closestPixel.(px)
    end

    return hcat(R...) # convert our array of arrays to a single matrix of pixels
end

Original image (1920x1080):

In MC 1.8: