https://github.com/benmaier/kd-tree-js
A JavaScript implementation of a kd tree that enables fast searches for points within a range as well as nearest neighbors in d dimensions
https://github.com/benmaier/kd-tree-js
Last synced: 3 months ago
JSON representation
A JavaScript implementation of a kd tree that enables fast searches for points within a range as well as nearest neighbors in d dimensions
- Host: GitHub
- URL: https://github.com/benmaier/kd-tree-js
- Owner: benmaier
- License: mit
- Created: 2022-11-16T16:17:02.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2022-11-16T19:04:48.000Z (over 2 years ago)
- Last Synced: 2024-10-11T11:06:40.717Z (8 months ago)
- Language: JavaScript
- Homepage:
- Size: 6.84 KB
- Stars: 4
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# kd-tree-js
A JavaScript implementation of a kd tree that enables fast searches for nearest neighbors in d dimensions.
Supports range searches (finding points within ball R) and k-nearest-neighbor searches.
```javascript
let tree = new KDTree(points, ['x','y']);
let query_point = {x:0.2,y:0.3};
let R = 0.2;
let ball = tree.nearest_within_R(query_point, R);
let nearest_k = tree.nearest_k(query_point, 3);
let nearest_neighbor = tree.nearest(query_point);
```
`ball` will be an ordered list of objects. Each object is structured like `{point: {..., index: 9}, squared_distance: 0.02}` and orderd by squared distance. `nearest_k` has the same structure. `nearest_neighbor` is just the first element of the list from `nearest_k` for `k>0`.
## How to use
### p5 example
[p5 demo](https://editor.p5js.org/kimxsegfault/sketches/0--CV4nQn).
https://user-images.githubusercontent.com/10728380/202268588-79b53ec1-4c96-4831-aac2-9304eb4fc501.mov
```javascript
let points = [];
let N = 2000;
let tree;
let r = 3;
function setup() {
createCanvas(400, 400);
for(let i=0; i{
circle(p.x,p.y,2*r);
})
let R = 50;
let k = 50;
push();
noFill();
strokeWeight(2);
circle(mouseX, mouseY, R*2);
pop()
push();
let ball = tree.nearest_within_R({x:mouseX,y:mouseY},R);
let nearest_k = tree.nearest_k({x:X,y:Y},k);
fill(180);
ball.forEach(n=>{
circle(n.point.x,n.point.y,2*r);
})
fill(100);
nearest_k.forEach(n=>{
circle(n.point.x,n.point.y,2*r);
})
pop();
}
```
### 2D example
```javascript
let points = [
{x: 0.5, y:0.5},
{x: 0.1, y:0.3},
{x: 1, y:0.1},
{x: 0.75, y:2.5},
{x: 0.7, y:1.2}
];
let tree = new KDTree(points, ['x','y']);
let query_point = {x:0.2,y:0.3};
let R = 0.2;
let ball = tree.nearest_within_R(query_point, R);
let nearest_k = tree.nearest_k(query_point, 3);
let nearest_neighbor = tree.nearest(query_point);
```
```javascript
> ball
[ // ordered by squared distance
{
point: {
x: 0.1,
y: 0.3,
idx: 1
},
squared_distance: 0.01
}
]
```
```javascript
> nearest_k
[ // ordered by squared distance
{
point: {
x: 0.1,
y: 0.3,
idx: 1
},
squared_distance: 0.01
},
{
point: {
x: 0.5,
y: 0.5,
idx: 0
},
squared_distance: 0.13
},
{
point: {
x: 1,
y: 0.1,
idx: 2
},
squared_distance: 0.68
}
]
```
```javascript
> nearest_neighbor
{
point: {
x: 0.1,
y: 0.3,
idx: 1
},
squared_distance: 0.01
}
```
### 3D example
```javascript
let points = [
[ 0.5, 0.5, 0.6],
[ 0.4, 0.2, 0.8],
[ 0.1, 0.7, 0.1],
[ 0.3, 0.7, 0.2]
];
let tree = new KDTree(points, [0,1,2]);
let nearest_neighbor = tree.nearest([0.5,0.5,0.5]);
console.log(nearest_neighbor);
```
Console:
```javascript
{
point: {
0: 0.5,
1: 0.5,
2: 0.6,
idx: 0
},
squared_distance: 0.01
}
```