Supervised Classification (Table Fields)

Author: O.Conrad (c) 2012
Menu: Table | Classification;Shapes | Attributes

Description

Standard classifiers for supervised classification based on attributes. Classifiers can be trained in three different ways:

Known classes field: choose an attribute field that provides class identifiers for those records, for which the target class is known, and no-data (or empty string) for all other records.
Training samples: a table with sample records providing the class identifier in the first field followed by sample data corresponding to the selected feature attributes.
Load statistics from file: loads feature statistics from a file that has been previously stored after training with one of the other two options.

Parameters

	Name	Type	Identifier	Description	Constraints
Input	Table	table, input	`TABLE`	-	-
Input	Training Samples	table, input	`TRAIN_SAMPLES`	Provide a class identifier in the first field followed by sample data corresponding to the selected feature attributes.	-
Output	Classification	table, output, optional	`RESULT_TABLE`	-	-
Output	Classification	shapes, output, optional	`RESULT_SHAPES`	-	-
Options	Features	table fields	`FEATURES`	-	-
	Normalise	boolean	`NORMALISE`	-	Default: 0
	Training	choice	`TRAIN_WITH`	-	Available Choices: [0] known classes field [1] training samples [2] load statistics from file Default: 0
	Known Classes Field	table field	`TRAIN_FIELD`	-	-
	Load Statistics from File...	file path	`FILE_LOAD`	-	-
	Save Statistics to File...	file path	`FILE_SAVE`	-	-
	Method	choice	`METHOD`	-	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 Threshold	floating point number	`THRESHOLD_DIST`	Let pixel stay unclassified, if minimum euclidean or mahalanobis distance is greater than threshold.	Minimum: 0.000000 Default: 0.000000
	Spectral Angle Threshold (Degree)	floating point number	`THRESHOLD_ANGLE`	Let pixel stay unclassified, if spectral angle distance is greater than threshold.	Minimum: 0.000000 Maximum: 90.000000 Default: 0.000000
	Probability Threshold	floating point number	`THRESHOLD_PROB`	Let pixel stay unclassified, if maximum likelihood probability value is less than threshold.	Minimum: 0.000000 Maximum: 100.000000 Default: 0.000000
	Probability Reference	choice	`RELATIVE_PROB`	-	Available Choices: [0] absolute [1] relative Default: 1
	Binary Encoding	boolean	`WTA_0`	-	Default: 0
	Parallelepiped	boolean	`WTA_1`	-	Default: 0
	Minimum Distance	boolean	`WTA_2`	-	Default: 0
	Mahalanobis Distance	boolean	`WTA_3`	-	Default: 0
	Maximum Likelihood	boolean	`WTA_4`	-	Default: 0
	Spectral Angle Mapping	boolean	`WTA_5`	-	Default: 0

Command Line


Usage: saga_cmd table_tools 26 [-TABLE ] [-FEATURES ] [-NORMALISE ] [-RESULT_TABLE ] [-RESULT_SHAPES ] [-TRAIN_WITH ] [-TRAIN_FIELD ] [-TRAIN_SAMPLES ] [-FILE_LOAD ] [-FILE_SAVE ] [-METHOD ] [-THRESHOLD_DIST ] [-THRESHOLD_ANGLE ] [-THRESHOLD_PROB ] [-RELATIVE_PROB ] [-WTA_0 ] [-WTA_1 ] [-WTA_2 ] [-WTA_3 ] [-WTA_4 ] [-WTA_5 ]
  -TABLE:                Table
	table, input
  -FEATURES:             Features
	table fields
  -NORMALISE:            Normalise
	boolean
	Default: 0
  -RESULT_TABLE:         Classification
	table, output, optional
  -RESULT_SHAPES:        Classification
	shapes, output, optional
  -TRAIN_WITH:
                              Training
                              	choice
                              	Available Choices:
                              	[0] known classes field
                              	[1] training samples
                              	[2] load statistics from file
                              	Default: 0
  -TRAIN_FIELD:          Known Classes Field
	table field
  -TRAIN_SAMPLES:        Training Samples
	table, input
  -FILE_LOAD:            Load Statistics from File...
	file path
  -FILE_SAVE:            Save Statistics to File...
	file path
  -METHOD:
                              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:    Distance Threshold
	floating point number
	Minimum: 0.000000
	Default: 0.000000
  -THRESHOLD_ANGLE:
                              Spectral Angle Threshold (Degree)
                              	floating point number
                              	Minimum: 0.000000
                              	Maximum: 90.000000
                              	Default: 0.000000
  -THRESHOLD_PROB:
                              Probability Threshold
                              	floating point number
                              	Minimum: 0.000000
                              	Maximum: 100.000000
                              	Default: 0.000000
  -RELATIVE_PROB:        Probability Reference
	choice
	Available Choices:
	[0] absolute
	[1] relative
	Default: 1
  -WTA_0:                Binary Encoding
	boolean
	Default: 0
  -WTA_1:                Parallelepiped
	boolean
	Default: 0
  -WTA_2:                Minimum Distance
	boolean
	Default: 0
  -WTA_3:                Mahalanobis Distance
	boolean
	Default: 0
  -WTA_4:                Maximum Likelihood
	boolean
	Default: 0
  -WTA_5:                Spectral Angle Mapping
	boolean
	Default: 0