SAGA-GIS Tool Library Documentation (v6.1.0)

Tool Supervised Classification for Shapes

Supervised classification for attribute data. To train the classifier choose an attribute that provides class identifiers for those records, for which the target class is known, and no data for all other records.

Parameters

 NameTypeIdentifierDescriptionConstraints
InputShapesShapes (input)SHAPES--
OutputClassificationShapes (output)CLASSES--
OptionsFeaturesTable fieldsFEATURES--
NormaliseBooleanNORMALISE-Default: 0
Training Classes (*)Table fieldTRAINING--
Load Statistics from File...File pathFILE_LOAD--
Save Statistics to File...File pathFILE_SAVE--
MethodChoiceMETHOD-Available Choices:
[0] Binary Encoding
[1] Parallelepiped
[2] Minimum Distance
[3] Mahalanobis Distance
[4] Maximum Likelihood
[5] Spectral Angle Mapping
[6] Winner Takes All
Default: 2
Distance ThresholdFloating pointTHRESHOLD_DISTLet pixel stay unclassified, if minimum euclidian or mahalanobis distance is greater than threshold.Minimum: 0.000000
Default: 0.000000
Spectral Angle Threshold (Degree)Floating pointTHRESHOLD_ANGLELet pixel stay unclassified, if spectral angle distance is greater than threshold.Minimum: 0.000000
Maximum: 90.000000
Default: 0.000000
Probability ThresholdFloating pointTHRESHOLD_PROBLet pixel stay unclassified, if maximum likelihood probability value is less than threshold.Minimum: 0.000000
Maximum: 100.000000
Default: 0.000000
Probability ReferenceChoiceRELATIVE_PROB-Available Choices:
[0] absolute
[1] relative
Default: 1
Binary EncodingBooleanWTA_0-Default: 0
ParallelepipedBooleanWTA_1-Default: 0
Minimum DistanceBooleanWTA_2-Default: 0
Mahalanobis DistanceBooleanWTA_3-Default: 0
Maximum LikelihoodBooleanWTA_4-Default: 0
Spectral Angle MappingBooleanWTA_5-Default: 0
(*) optional

Command-line

Usage: saga_cmd imagery_classification 4 [-SHAPES <str>] [-CLASSES <str>] [-FEATURES <str>] [-NORMALISE <str>] [-TRAINING <str>] [-FILE_LOAD <str>] [-FILE_SAVE <str>] [-METHOD <str>] [-THRESHOLD_DIST <double>] [-THRESHOLD_ANGLE <double>] [-THRESHOLD_PROB <double>] [-RELATIVE_PROB <str>] [-WTA_0 <str>] [-WTA_1 <str>] [-WTA_2 <str>] [-WTA_3 <str>] [-WTA_4 <str>] [-WTA_5 <str>]
  -SHAPES:<str>            	Shapes
	Shapes (input)
  -CLASSES:<str>           	Classification
	Shapes (output)
  -FEATURES:<str>          	Features
	Table fields
  -NORMALISE:<str>         	Normalise
	Boolean
	Default: 0
  -TRAINING:<str>          	Training Classes
	Table field
  -FILE_LOAD:<str>         	Load Statistics from File...
	File path
  -FILE_SAVE:<str>         	Save Statistics to File...
	File path
  -METHOD:<str>            	Method
	Choice
	Available Choices:
	[0] Binary Encoding
	[1] Parallelepiped
	[2] Minimum Distance
	[3] Mahalanobis Distance
	[4] Maximum Likelihood
	[5] Spectral Angle Mapping
	[6] Winner Takes All
	Default: 2
  -THRESHOLD_DIST:<double> 	Distance Threshold
	Floating point
	Minimum: 0.000000
	Default: 0.000000
  -THRESHOLD_ANGLE:<double>	Spectral Angle Threshold (Degree)
	Floating point
	Minimum: 0.000000
	Maximum: 90.000000
	Default: 0.000000
  -THRESHOLD_PROB:<double> 	Probability Threshold
	Floating point
	Minimum: 0.000000
	Maximum: 100.000000
	Default: 0.000000
  -RELATIVE_PROB:<str>     	Probability Reference
	Choice
	Available Choices:
	[0] absolute
	[1] relative
	Default: 1
  -WTA_0:<str>             	Binary Encoding
	Boolean
	Default: 0
  -WTA_1:<str>             	Parallelepiped
	Boolean
	Default: 0
  -WTA_2:<str>             	Minimum Distance
	Boolean
	Default: 0
  -WTA_3:<str>             	Mahalanobis Distance
	Boolean
	Default: 0
  -WTA_4:<str>             	Maximum Likelihood
	Boolean
	Default: 0
  -WTA_5:<str>             	Spectral Angle Mapping
	Boolean
	Default: 0