0
$\begingroup$

Assume that you having a heatmap that looks like this. The goal is to classify all the "dot" inside the image. How can that be done?

The assumptions of the image:

  • The image has always black background
  • The white is values that are around 100 or more.
  • In each cluster, there are only one maximum value

enter image description here

I want to know

  • How many clusters there are
  • The radius of the clusters

To do so, I need to separate them into different clusters. But how can that be done?

Improve this question
$\endgroup$
4
  • $\begingroup$ It depends on what you really mean by a "cluster", why you want to do this, and the expected geometries of the clusters. For instance, I describe and illustrate a workflow for clustering a Hough transform at stats.stackexchange.com/a/581400/919. Although it works well for that form of data, it wouldn't necessarily be a good idea in other applications, which is why these details matter. See, for instance, mathematica.stackexchange.com/questions/20696 for a rather different approach. $\endgroup$
    – whuber ♦
    Commented Sep 11, 2023 at 20:49
  • 1
    $\begingroup$ @whuber Thank you for your reply. This plot is actually a hough transform. I'm seeking advice how to improve the findings of the peaks. $\endgroup$
    – euraad
    Commented Sep 11, 2023 at 20:58
  • $\begingroup$ Have you tried the solutions I linked to? $\endgroup$
    – whuber ♦
    Commented Sep 13, 2023 at 14:41
  • 1
    $\begingroup$ @whuber I solved this issue by DBscan, on x-axis and y-axis only. [x, y] = find(A) and make a cluster identification using dbscan onto matrix [x y] $\endgroup$
    – euraad
    Commented Sep 13, 2023 at 15:58

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.