MATLAB Code for K-Harmonic Means clustering

16 views (last 30 days)
su mon aung
su mon aung on 31 Oct 2015
Answered: Bernhard Suhm on 24 Dec 2017
Hi , how I can write Matlab code for K-Harmonic Means clustering to extract feature from image background.I respectfully request to reply my question as soon as possible.
  2 Comments
D.Regan
D.Regan on 18 Nov 2017
hi..! This is the paper in which the implementation of K-Harmonic Means available,"Spatial kernel K-harmonic means clustering for multi-spectral image segmentation".

Sign in to comment.

Answers (1)

Bernhard Suhm
Bernhard Suhm on 24 Dec 2017
This algorithm is not supported in the Statistics and Machine Learning toolbox, only standard k-means with various distance functions. However you could try to use regular K-Means to do this. You will have to carefully select the feature space to cluster in as 3x image dimensions are likely too big and noisy.
Our usual recommendation is to use the Image Segmented app from the Image Processing toolbox and try a few segmentation techniques that use more sophisticated clustering under the hood. If that doesn’t work you could try using extractFeatures from the Computer Vision toolbox.

Categories

Find more on Statistics and Machine Learning Toolbox in Help Center and File Exchange

Community Treasure Hunt

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

Start Hunting!

Translated by