Clusterig points in clusters, separated by a minimum distance

Question

Alexamder Schweizer 2019-4-19

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/457352-clusterig-points-in-clusters-separated-by-a-minimum-distance

回答： Akshat 2024-9-4，9:26

I have a set of 300 points, defined by x and y coordinates in separat vectors. I want to cluster the points in such a way, that every center of the cluster is at least a defined distance from the closest other one. The distance can not be less than the defined one.

Is there a way that I dont have to define the exact number of clusters?

Which function should I use?

Thanks in advance for the response

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Akshat 2024-9-4，9:26

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/457352-clusterig-points-in-clusters-separated-by-a-minimum-distance#answer_1510549

在 MATLAB Online 中打开

Hi Alexamder,

The functionality you are asking for, which involves clustering points with a threshold for a neighborhood search radius, is provided in MATLAB using "dbscan".

Here is the documentation for the same:

https://www.mathworks.com/help/stats/dbscan.html

I have written a sample code with 300 points for you to see how the function works, here:

x = rand(300, 1) * 100; 
y = rand(300, 1) * 100; 
data = [x, y];
epsilon = 5; % Minimum distance between cluster centers
min_samples = 5; % Minimum number of points to form a cluster
idx = dbscan(data, epsilon, min_samples);
figure;
gscatter(data(:,1), data(:,2), idx);
xlabel('X');
ylabel('Y');
grid on;