DESCRIPTION

This is the KMeansClassification application, version 5.2.0 Unsupervised KMeans image classification

Complete documentation: http://www.orfeo-toolbox.org/Applications/KMeansClassification.html

Parameters:

-progress <boolean>

Report progress

 -in       <string>         Input Image  (mandatory)
 -out      <string> [pixel] Output Image  [pixel=uint8/uint16/int16/uint32/int32/float/double] (default value is uint8) (mandatory)

-ram

<int32> Available RAM (Mb) (optional, off by default, default value is 128)

-vm

<string> Validity Mask (optional, off by default)

-ts

<int32> Training set size (optional, on by default, default value is 100)

-nc

<int32> Number of classes (mandatory, default value is 5)

-maxit

<int32> Maximum number of iterations (optional, on by default, default value is 1000)

-ct

<float> Convergence threshold (optional, on by default, default value is 0.0001)

-outmeans <string>

Centroid filename (optional, off by default)

-rand

<int32> set user defined seed (optional, off by default)

-inxml

<string> Load otb application from xml file (optional, off by default)

EXAMPLES

otbcli_KMeansClassification -in QB_1_ortho.tif -ts 1000 -nc 5 -maxit 1000 -ct 0.0001 -out ClassificationFilterOutput.tif