KD Tree & KNN [2D]

Overview

As you know, KDTREE is an algorithm that efficiently sorts large amounts of data, and KNN (KNN) is the ability to search for a specific range of data using KD Tree.

This is an example code, which uses a recursive function to sort the data. We also leverage the Eucladian distance to run KNN.

Please look at it with the comments in the main and make it useful. Thank you.

---

---

build

$> cd KD-Tree_and_KNN
$> mkdir build && cd build
$> cmake .. && make
$> ./kdtree

---

Select Senario

Choose Senario. 0=Sample, 1=Pcd

_

Senario 0

▶ 1. You must choose Width and Height for Point Area [Width, Height]

▶ 2. Write the number of Points to create

▶ 3. Write the Root position [x, y] for find nearest point

▶ 4. Write K Point whatever you want. This is the number of Near position data of Root position

_

Senario 1

▶ 1. Write the Root position [x, y] for find nearest point

▶ 2. Write K Point whatever you want. This is the number of Near position data of Root position

Sample.pcd

CornerMap.pcd

---

Helped with the reference.

[https://github.com/gishi523/kd-tree]