Knnsearch: How to output disntace?

Question

Pelajar UM 2021-10-26

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1571928-knnsearch-how-to-output-disntace

编辑： Pelajar UM 2021-10-26

Here's the code. It basically finds the nearest sets of x,y,z points in the nodes array. Since the first column is the point itself, K=2, so that it finds the second nearest point. Then it generates a new array Near which sorts the data based on this new index.

Idx=knnsearch (nodes,nodes,'K',2)
Idx2=Idx(:,2)
Near=nodes(Idx2,:)

According to the documentation, there's an output argument D that should give the distance. But how do I implement it?

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

Esen Ozbay 2021-10-26

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1571928-knnsearch-how-to-output-disntace#answer_816748

If you write

[Idx, dist] =knnsearch (nodes,nodes,'K',2);

you will obtain the distances mentioned in the documentation.

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Pelajar UM 2021-10-26

编辑：Pelajar UM 2021-10-26

Perfect. Thanks a lot!

Although some of the distances don't make a lot of sense. I'm trying to find the thickness of a 3D object. The generatemesh function was used to create nodes from an stl file. But the position of the nodes are quite random so I don't think Euclidean distance is the best approach for this problem.

请先登录，再进行评论。