`dsearchn`

Index of the nearest point in a set, for each query point.

Category: Mathematics > Computational Geometry

Syntax

k = dsearchn(P, PQ)

For each row of PQ, finds the closest row of P.

[k, d] = dsearchn(P, PQ)

Also returns the distances.

Description

The distance is the ordinary Euclidean distance. The search compares each query point against the data points, so the result is exact; if two data points are equally close, one of them is returned.

The function works in any dimension and is a quick way to snap points onto a mesh or a sample set.

Input arguments

P: Matrix of data points, one per row, with n columns for n dimensions.
PQ: Query points, one per row, with the same number of columns as P.

Output arguments

k: Column vector with, for each query point, the row index in P of its nearest neighbor.
d: Column vector of the Euclidean distances to those neighbors.

Example

Nearest of three points

P = [0 0; 5 5; 10 0];
PQ = [4 4; 9 1; 1 0];
[k, d] = dsearchn(P, PQ)

Output

dsearchn