Tool NameRandom Forest Classification
Tool ID10
Library IDimagery_opencv
Version1.0
Author(s)O.Conrad (c) 2016

Description

Integration of the OpenCV Machine Learning library for Random Forest classification of gridded features.

References

Breiman, L. (2001): Random Forests. Machine Learning, 45 (1): 5-32. doi:10.1023/A:1010933404324

OpenCV - Open Source Computer Vision

OpenCV - Machine Learning Overview


Parameters

NameTypeIdentifierDescriptionConstraints
Input
Featuresgrid list, inputFEATURES
Training Samplestable, inputTRAIN_SAMPLESProvide a class identifier in the first field followed by sample data corresponding to the input feature grids.
Training Areasshapes, inputTRAIN_AREAS
Output
Classificationgrid, outputCLASSES
Look-up Tabletable, output, optionalCLASSES_LUTA reference list of the grid values that have been assigned to the training classes.
Feature Importancetable, output, optionalIMPORTANCE
Options
NormalizebooleanNORMALIZEDefault: 0
TrainingchoiceMODEL_TRAINAvailable Choices:
[0] training areas
[1] training samples
[2] load from file
Default: 0
Class Identifiertable fieldTRAIN_CLASS
Buffer Sizefloating point numberTRAIN_BUFFERFor non-polygon type training areas, creates a buffer with a diameter of specified size.Minimum: 0.000000
Default: 1.000000
Load Modelfile pathMODEL_LOADUse a model previously stored to file.
Save Modelfile pathMODEL_SAVEStores model to file to be used for subsequent classifications instead of training areas.
Maximum Tree Depthinteger numberMAX_DEPTHThe maximum possible depth of the tree. That is the training algorithms attempts to split a node while its depth is less than maxDepth. The root node has zero depth.Minimum: 1
Default: 10
Minimum Sample Countinteger numberMIN_SAMPLESIf the number of samples in a node is less than this parameter then the node will not be split.Minimum: 2
Default: 2
Maximum Categoriesinteger numberMAX_CATEGRSCluster possible values of a categorical variable into K<=maxCategories clusters to find a suboptimal split.Minimum: 1
Default: 10
Use 1SE Ruleboolean1SE_RULEIf true then a pruning will be harsher. This will make a tree more compact and more resistant to the training data noise but a bit less accurate.Default: 1
Truncate Pruned TreesbooleanTRUNC_PRUNEDIf true then pruned branches are physically removed from the tree. Otherwise they are retained and it is possible to get results from the original unpruned (or pruned less aggressively) tree.Default: 1
Regression Accuracyfloating point numberREG_ACCURACYTermination criteria for regression trees. If all absolute differences between an estimated value in a node and values of train samples in this node are less than this parameter then the node will not be split further.Minimum: 0.000000
Default: 0.010000
Active Variable Countinteger numberACTIVE_VARSThe size of the randomly selected subset of features at each tree node and that are used to find the best split(s). If you set it to 0 then the size will be set to the square root of the total number of features.Minimum: 0
Default: 0
Number of Treesinteger numberNTREESThe number of decision trees to be built in the forest.Minimum: 0
Default: 5