Cluster 3D data

25 次查看（过去 30 天）
Amarbold Purev
Amarbold Purev 2021-2-23
评论： darova 2021-2-24
Hello,
I have a combined X Y Z camera coordinates and I want to cluster them into 2 groups.
I have tried the following code:
n = length(camLoc);
cam_A = NaN(n,3);
cam_B = NaN(n,3);
idx = kmeans(camLoc,2, 'Distance', 'cityblock', 'Start', 'uniform', 'Replicates',5);
for i = 1:n
if idx(i) == 1
cam_A(i,:) = camLoc(i,:);
else
cam_B(i,:) = camLoc(i,:);
end
end
cam_A(any(isnan(cam_A), 2), :) = [];
cam_B(any(isnan(cam_B), 2), :) = [];
scatter3(cam_A(:,1),cam_A(:,2),cam_A(:,3),'r');
hold on
scatter3(cam_B(:,1),cam_B(:,2),cam_B(:,3),'b');
and the results is shown below.
However, I want to cluster not vertically but horizontally. In other words, I want the data, that it would look like a 2 long noodles.
Thanks in advance.

请先登录，再回答此问题。

采纳的回答

darova
darova 2021-2-23
  2 个评论
Amarbold Purev
Amarbold Purev 2021-2-24
Thank you @darova it really helped.
Simple clusterdata worked like a charm.
idx = clusterdata(camLoc,2);

请先登录，再进行评论。

更多回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Point Cloud Processing 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by