A nice and fast way to construct 2D Voronoi diagrams written in Rust.
Voronoice builds Voronoi diagrams by first obtaining its Delaunay triangulation, through the really fast delaunator crate and then extracting its dual Voronoi diagram.
```rust use voronoice::*;
// voronoi sites let sites = vec![ Point { x: 0.0, y: 0.0 }, Point { x: 1.0, y: 0.0 }, Point { x: 0.0, y: 1.0 } ];
// builds a voronoi diagram from the set of sites above, bounded by a square of size 4 let myvoronoi = VoronoiBuilder::default() .setsites(sites) .setboundingbox(BoundingBox::newcenteredsquare(4.0)) .setlloydrelaxation_iterations(5) .build() .unwrap();
// inspect cells through iterators
myvoronoi.itercells().for_each(|cell| {
println!("Vertices of cell: {:?}", cell.vertices().collect::
// or probe cells individually
let mycell = myvoronoi.cell(1);
println!("Second cell has site {:?}, voronoi vertices {:?} and delaunay triangles {:?}",
mycell.siteposition(),
mycell.vertices().collect::
// or, for graphical applications, that benefit from index buffers // you can access the raw, indexed data let allvoronoicellvertices = myvoronoi.vertices(); let indexedvoronoicells = myvoronoi.cells(); println!("The first vertex position for the first voronoi cell is at {:?}", allvoronoicellvertices[indexedvoronoicells[0][0]]); ```
On docs.rs.
Here are some generation times on a 3.5GHz Core i7 from 2012.
| Number of points | Time | | -----------------|--------------| | 1,000 | 150 µs | | 10,000 | 1.5 ms | | 100,000 | 18 ms | | 1,000,000 | 270 ms | | 5,000,000 | 1.6 s | | 10,000,000 | 3.5 s |
A comparison benchmark with other Rust Voronoi diagram generating libraries can be found here.
cargo run --example svg -- -s10 -l2 -z0.8 -o example.svg --render-labels false
Generates a SVG image to visualize the Voronoi-Delaunay graph. Provide --help to see other options.
cargo run --release --example image -- examples/assets/mona_noice_small.jpg 300
Generates a color averaged image by overlaying a Voronoi diagram
