Struct ConvexHull 

pub struct ConvexHull { /* private fields */ }

A ConvexHull represents the convex hull of a set of points.

It provides methods to access hull properties, check if points are inside the hull, and access hull facets and vertices.

Implementations

impl ConvexHull

pub fn new(points: &ArrayView2<'_, f64>) -> SpatialResult<Self>

Create a new ConvexHull from a set of points.

Arguments

• points - Input points (shape: npoints x n_dim)

Returns

• Result containing a ConvexHull instance or an error

Errors

• Returns error if hull computation fails or input is invalid

Examples

use scirs2_spatial::convex_hull::ConvexHull;
use scirs2_core::ndarray::array;

let points = array![[0.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.5, 0.5]];
let hull = ConvexHull::new(&points.view()).unwrap();

pub fn new_with_algorithm( points: &ArrayView2<'_, f64>, algorithm: ConvexHullAlgorithm, ) -> SpatialResult<Self>

Create a new ConvexHull from a set of points using a specific algorithm.

Arguments

• points - Input points (shape: npoints x n_dim)
• algorithm - Algorithm to use for convex hull computation

Returns

• Result containing a ConvexHull instance or an error

Errors

• Returns error if hull computation fails or input is invalid
• Returns error if algorithm is not supported for the given dimensionality

Examples

use scirs2_spatial::convex_hull::{ConvexHull, ConvexHullAlgorithm};
use scirs2_core::ndarray::array;

let points = array![[0.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.5, 0.5]];
let hull = ConvexHull::new_with_algorithm(&points.view(), ConvexHullAlgorithm::GrahamScan).unwrap();

pub fn vertices(&self) -> Vec<Vec<f64>>

Get the vertices of the convex hull

Returns

• Array of vertices of the convex hull

Examples

use scirs2_spatial::convex_hull::ConvexHull;
use scirs2_core::ndarray::array;

let points = array![[0.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.5, 0.5]];
let hull = ConvexHull::new(&points.view()).unwrap();

// The hull vertices should be the corners, not the interior point
// The number of vertices can vary depending on QHull implementation
assert!(hull.vertices().len() >= 3);

pub fn vertices_array(&self) -> Array2<f64>

Get the vertices of the convex hull as an Array2

Returns

• Array2 of shape (n_vertices, n_dim) containing the vertices of the convex hull

pub fn vertex_indices(&self) -> &[usize]

Get the indices of the vertices of the convex hull

Returns

• Vector of indices of the hull vertices (into the original points array)

Examples

use scirs2_spatial::convex_hull::ConvexHull;
use scirs2_core::ndarray::array;

let points = array![[0.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.5, 0.5]];
let hull = ConvexHull::new(&points.view()).unwrap();

// Get the vertex indices of the hull
let vertices = hull.vertex_indices();
println!("Convex hull vertices: {:?}", vertices);

pub fn simplices(&self) -> &[Vec<usize>]

Get the simplices (facets) of the convex hull

Returns

• Vector of simplices, where each simplex is a vector of vertex indices

Examples

use scirs2_spatial::convex_hull::ConvexHull;
use scirs2_core::ndarray::array;

let points = array![[0.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.5, 0.5]];
let hull = ConvexHull::new(&points.view()).unwrap();

// For a 2D hull, examine the simplices
for simplex in hull.simplices() {
    // Print each simplex
    println!("Simplex: {:?}", simplex);
}

pub fn ndim(&self) -> usize

Get the dimensionality of the convex hull

Returns

• Number of dimensions of the convex hull

Examples

use scirs2_spatial::convex_hull::ConvexHull;
use scirs2_core::ndarray::array;

let points = array![[0.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.5, 0.5]];
let hull = ConvexHull::new(&points.view()).unwrap();

assert_eq!(hull.ndim(), 2);

pub fn contains<T: AsRef<[f64]>>(&self, point: T) -> SpatialResult<bool>

Check if a point is inside the convex hull

Arguments

• point - The point to check

Returns

• Result containing a boolean indicating if the point is inside the hull

Errors

• Returns error if point dimension doesn’t match hull dimension

Examples

use scirs2_spatial::convex_hull::ConvexHull;
use scirs2_core::ndarray::array;

let points = array![[0.0, 0.0], [1.0, 0.0], [0.0, 1.0]];
let hull = ConvexHull::new(&points.view()).unwrap();

// Check a point inside the hull
// Result may vary depending on QHull implementation and special case handling
let inside = hull.contains(&[0.25, 0.25]).unwrap();
println!("Point [0.25, 0.25] inside: {}", inside);

// Check a point outside the hull
assert!(!hull.contains(&[2.0, 2.0]).unwrap());

pub fn volume(&self) -> SpatialResult<f64>

Get the volume of the convex hull

For 2D hulls, this returns the area. For 3D hulls, this returns the volume. For higher dimensions, this returns the hypervolume.

Returns

• Result containing the volume/area of the convex hull

Errors

• Returns error if volume computation fails

Examples

use scirs2_spatial::convex_hull::ConvexHull;
use scirs2_core::ndarray::array;

// Create a square with area 1
let points = array![[0.0, 0.0], [1.0, 0.0], [1.0, 1.0], [0.0, 1.0]];
let hull = ConvexHull::new(&points.view()).unwrap();

let area = hull.volume().unwrap();
assert!((area - 1.0).abs() < 1e-10);

pub fn area(&self) -> SpatialResult<f64>

Get the surface area of the convex hull

For 2D hulls, this returns the perimeter. For 3D hulls, this returns the surface area. For higher dimensions, this returns the surface hypervolume.

Returns

• Result containing the surface area/perimeter of the convex hull

Errors

• Returns error if area computation fails

Examples

use scirs2_spatial::convex_hull::ConvexHull;
use scirs2_core::ndarray::array;

// Create a 3D cube
let points = array![
    [0.0, 0.0, 0.0], [1.0, 0.0, 0.0], [1.0, 1.0, 0.0], [0.0, 1.0, 0.0],
    [0.0, 0.0, 1.0], [1.0, 0.0, 1.0], [1.0, 1.0, 1.0], [0.0, 1.0, 1.0]
];
let hull = ConvexHull::new(&points.view()).unwrap();

// Surface area of the cube should be 6 square units
let area = hull.area().unwrap();
assert!((area - 6.0).abs() < 1e-10);

pub fn extract_equations(qh: &Qh<'_>, ndim: usize) -> Option<Array2<f64>>

Extract facet equations from the Qhull instance

Arguments

• qh - Qhull instance
• ndim - Dimensionality of the hull

Returns

• Option containing an Array2 of shape (n_facets, ndim+1) with the equations of the hull facets or None if equations cannot be extracted