Implementing LBG algorithm

New to Julia
1

Hi, I’m new to Julia and wondering how to loop through the generated voronoi cells to implement the MLBG algorithm.


using GLMakie
using DelaunayTriangulation
using Interpolations
using Plots


function perform_computations(axis, time_range, mean_data, std_data, mean_color, fig)
    # band!(time_range, mean_data - std_data, mean_data + std_data, color=GLMakie.RGBA(mean_color, 0.6), label="Error")

    lines!(axis, time_range, mean_data, color=GLMakie.RGBA(mean_color, 1.0), label="Mean")
    # xlims!(axis, [0, 50])
    time_range_dense = range(1, stop=length(time_range), length=length(time_range) * 20)

    itp_mean = interpolate((time_range,), mean_data, Gridded(Linear()))

    itp_lower = interpolate((time_range,), mean_data - std_data, Gridded(Linear()))
    itp_upper = interpolate((time_range,), mean_data + std_data, Gridded(Linear()))

    plot_ys_mean = [itp_mean(t) for t in time_range_dense]
    plot_ys_lower = [itp_lower(t) for t in time_range_dense]
    plot_ys_upper = [itp_upper(t) for t in time_range_dense]
    ax = Axis(fig[1, 2])

    GLMakie.scatter!(axis, time_range_dense, plot_ys_mean, color=:blue, label="Interpolation", markersize=3)

    error_ranges = plot_ys_upper .- plot_ys_lower


    tri = DelaunayTriangulation.triangulate(points; rng)

    vorn = voronoi(tri)

end

end

I need to loop through each of voronoi cells check if the density is lesser and update

stippleList ← initialize stipple list repeat
vd ← computeVoronoiDiagram(stippleList) newStippleList ← empty list
for all Voronoi cell vc ∈ vd do
density ← vc.sumDensity if density < Tl then
// remove point
else if density < Tu then
// keep point
newStippleList.add(vc.centroid)
else
// split point
splitPoints ← splitCell(vc) newStippleList.add(splitPoints)
end if end for
stippleList ← newStippleList until no splitting or merging
endfunction

Can anyone help me with this

2

What is the MLBG algorithm? It’s not clear to me what you want to do exactly. And P.S. NaturalNeighbours.jl might be a more natural choice for your interpolation (if I’m reading your code right?) since it’s directly built on top of DelaunayTriangulation.jl.

3

MLBG is [LindeBuzoGrayStippling Algorithm]. I’m trying to create stipples for the error bands using voronoi cells in order to avoid overlap when there are multiple time series data. Do you know How i can loop through the voronoi cells?

4

The documentation has examples of iterating over the Voronoi cells: Voronoi Tessellations · DelaunayTriangulation.jl. If anything there is unclear feel free to let me know and I’ll also update the documentation to address it.

1 Like
5

@DanielVandH thank you very much!

1 Like