Module: K-Means clustering

Description:

Groups pixels into K clusters using the K-means algorithm. Each cluster has a centroid, initialised to values spaced regularly along the intensity range (0-255). At each iteration, it groups the pixels into the clusters to which its intensity is closest to its centroid. At the end of each iteration, the centroid is updated as being equal to the average intensity of all the pixels in the cluster. The process is repeated until stability.

Example of a 4-Means clustering.

Connections:

Image

[required]
From this port, the input image is retrieved. For now, the only input type that has been implemented is HxUniformScalarField3 and byte data type (i.e. greyscale image with 256 grays).

Ports:

K

Number of clusters.

Cluster

Pressing this button starts the computation.

Commands:

Additional options can be accessed when typing in the console KMeans_Clustering COMMAND_NAME. Typing the command again usually reverts back to original settings.

`verbose`

Displays timing information after the computation. Retype to hide info.

`create`

Runs the computation. Returns the name of the output, so it can be used in a script, such as set RESULT [KMeans_Clustering create].