Unsupervised classification

Unsupervised K-Means classification

• In the search engine of the Processing Toolbox, type kmeans and double click KMeansClassification of OTB.
• Specify a multispectral image as Input Image.
• Specify directory and name for the Output image.
• Set the Number of classes to 20
• Check the Training set size to 10000
• Output pixel type: uint8
• Click on Run.

• Load the resulting image into QGIS. It is single band file with 20 grey levels labeled from 0 to 19.
• Layer Properties --> Symbology --> Render type. Switch to Singleband pseudocolor and select a Color ramp (e.g. Spectral). Select the Mode Equal interval and set the number of classes to 20. Apply.