How to find clusters in binary 3D image?

2015 7 10
1 个回答
回答已采纳
更新时间：2015 7 16
9 次查看（30 天）

0 个投票

Hello
I have a binary 3D image, i.e. it contains only 0 and 1. Now I want to find all clusters of 1s (i.e. clusters of voxels containing only value 1). Finally for each cluster I should know the coordinates of the voxels belonging to that cluster.
How can this be done? Of course I can iterate over all voxels but the difficulty is to detect the clusters and extract all voxels inside the cluster.

请先登录，再回答此问题。

 采纳的回答

Image Analyst
Image Analyst 2015-7-10

0 个投票

Use bwlabeln() or bwconncomp(). If you really need the (x,y,z) coordinates of each voxel in a labeled blob, then you can call regionprops() and ask for PixelIdxList.

2 个评论
显示 无 隐藏 无

Sepp
Sepp 2015-7-16
编辑：Sepp 2015-7-16
Thank you very much for your answer.
The binary 3D image is of size 60 x 90 x 10. With the following Matlab code I'm searching for connected voxels of 1s in the image:
conn = 26;
CC = bwconncomp(img, conn);
stats = regionprops(CC,'pixellist');
The really weird thing is that stats returns coordinates which are outside of the 3D image.
For example for
voxels = stats(6).PixelList;
voxels(23,:)
I'm getting a 3D coordinate of (61,15,4).
What is wrong here? Regarding this issue I'm also concerned that the calculation of the connected components is wrong.
Here is the 3D image (as .mat file): https://www.dropbox.com/s/pw3yzbcx4uyhh9s/img.mat?dl=0
Image Analyst
Image Analyst 2015-7-16
You've made probably the most common mistake novice users make : confusing x and y with row and column. Remember, all matrices in MATLAB are indexed by row and column, NOT by x and y. So it's M(row, column) = M(y,x), and NOT M(x,y) because x is not the row - it's the column.
PixelList returns a x,y,z vector. You have a matrix that you say is 60x90x10 which I assume is rows by columns by slices. So the max of x is 90, the max of y is 60, and the max of Z is 10. So a coordinate of (x,y,z) = (61,15,4) is well within the range of your matrix bounds of (90,60,10).

请先登录，再进行评论。

更多回答（0 个）

类别

在 帮助中心 和 File Exchange 中查找有关 Hierarchical Clustering 的更多信息

产品

标签

Community Treasure Hunt

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

Start Hunting!

Translated by