2D distribution/histogram's center of mass

4 次查看(过去 30 天)
I have a bunch of 2D histograms showing the relative dependency of two variables X and Y, under several conditions. Each histogram is 7x7, is produced with imagesc(), and looks something like this:
I would like to compute a single (scalar) measure of central tendency for each of these distributions. Something like a center of mass, although I havent found anything like this in the Matlab documentation, and I don't know anything about the field of image analysis.
Also, ideally, I could show this graphically as well by perhaps highlighting the cell of each distribution that contains this center of mass, or better still, a region within a certain radius of the center of mass that contains 68.2% (one standard deviation) of the volume of the distribution.
Any help much appreciated.
  4 个评论
José-Luis
José-Luis 2016-7-5
You could fit it to a multivariate normal distribution, see mvregress(), see how well it fits, and the statistics of the distribution could give some insight into the problem.
z8080
z8080 2016-7-5
I'll try that, thank you very much for your help

请先登录,再进行评论。

回答(0 个)

类别

Help CenterFile Exchange 中查找有关 Histograms 的更多信息

Community Treasure Hunt

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

Start Hunting!

Translated by