Cluster Analysis for Point Clouds

Volker Wichmann (c) 2010, LASERDATA GmbH
Cluster Analysis for Point Clouds.

Module usage is different between SAGA GUI and SAGA CMD: With SAGA GUI you will get prompted to choose the attributes to use once you execute the module. With SAGA CMD you have to provide a string with the -ATTR_FIELDS parameter containing the field numbers of the attributes to use (separated by semicolon). Field numbers start with 1, e.g. -ATTR_FIELDS="4;5;7".

References:

This module is a port of the 'Cluster Analysis for Grids' module from the 'Imagery - Classification' module library, Copyright (C) 2003 by Olaf Conrad.

Iterative Minimum Distance:
- Forgy, E. (1965):
'Cluster Analysis of multivariate data: efficiency vs. interpretability of classifications',
Biometrics 21:768

Hill-Climbing:- Rubin, J. (1967):
'Optimal Classification into Groups: An Approach for Solving the Taxonomy Problem',
J. Theoretical Biology, 15:103-144

Parameters

Point Cloud
Input Point Cloud
Input
Result
Output Point Cloud
Output
Statistics
Output Table
Method
Choice
Available choices: Iterative Minimum Distance (Forgy 1965), Hill-Climbing (Rubin 1967), Combined Minimum Distance / Hillclimbing
Clusters
Integer
Number of clusters
Minimum: 2.0
Normalise
Boolean
Automatically normalise attributes by standard deviation before clustering.
Attribute Fields
Text
The numbers (starting from 1) of the fields to use for clustering, separated by semicolon, e.g. "4;5;7"