How to calculate the centroid / center of a patch?

Question

Sim 2022-3-1

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1661115-how-to-calculate-the-centroid-center-of-a-patch

编辑： Sim 2022-3-1

采纳的回答： David Hill

在 MATLAB Online 中打开

Hi, how can I calculate the centroid / center of a patch?

Maybe something like this?

[Xc,Yc] = centroid(patch('XData',X,'YData',Y))

2 个评论
显示无隐藏无

jessupj 2022-3-1

i have you tried polyshape instead of a patch? you're defining the patch there with x,y data, so you can use those data to define the right object to apply centroid to

Sim 2022-3-1

Thanks a lot @jessupj ! Indeed, @David Hill showed (I think at the same time with you) this solution :)

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

David Hill 2022-3-1

2
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1661115-how-to-calculate-the-centroid-center-of-a-patch#answer_907130

在 MATLAB Online 中打开

pgon=polyshape(X,Y);
[x,y]=centroid(pgon);

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。

Answer 2

Constantino Carlos Reyes-Aldasoro 2022-3-1

2
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1661115-how-to-calculate-the-centroid-center-of-a-patch#answer_907125

The data X,Y must have been previously calculated from some raw data, say a thresholded figure, it is from that data you can easily calculate centroids with regionprops

https://uk.mathworks.com/help/images/ref/regionprops.html

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Sim 2022-3-1

Thanks a lot @Constantino Carlos Reyes-Aldasoro, very kind!

(however I accepted the @David Hill solution since showing an example of code)

请先登录，再进行评论。

Answer 3

Image Analyst 2022-3-1

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1661115-how-to-calculate-the-centroid-center-of-a-patch#answer_907205

在 MATLAB Online 中打开

Depends on what you have. If you have just arbitrary (x,y) data gotten somehow, like through assignment, then you can create a polyshape with the polyshape() function, and then use centroid(), like @David Hill showed.

If you have a binary image (like @Constantino Carlos Reyes-Aldasoro assumed -- say from thresholding), and want the centroid of each region, you can use regionprops():

props = regionprops(binaryImage, 'Centroid');
xyCentroids = vertcat(props.Centroid)

If you'd like a tutorial on image segmentation that shows that, you can see my Image Segmentation Tutorial:

Image Analyst's File Exchange

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Sim 2022-3-1

编辑：Sim 2022-3-1

Yes, very true, it depends on what you have!

Yes, both replies from @David Hill, @Constantino Carlos Reyes-Aldasoro and @jessupj are very good, and it just depends on what data / scenario you have....

Thanks a lot for the Tutorial! Seems very interesting! I guess this one: