ENVIConfusionMatrix

This is a reference to an ENVIConfusionMatrix object, which contains a confusion matrix and classification accuracy metrics that indicate how well a classifier performed. A confusion matrix is helpful for comparing the predicted (classification) results with truth data.

In an ENVI confusion matrix, columns represent true classes, while rows represent the classifier's predictions. The matrix is square, with all correct classifications along the upper-left to lower-right diagonal.

Here are some examples of how to read this matrix:

• 2,385 values were correctly classified as Asphalt.
• Reading down the Concrete column, 4 values that should have been Concrete were classified as Asphalt, and 1 value was classified as Grass. These are omission errors, which are defined as the fraction of values that belong to a class but were predicted to be in a different class. Errors of omission represent false negatives.
• Reading across the Concrete row, 1 value that should have been Building was classified as Concrete. This is a commission error, which is defined as the fraction of values that were predicted to be in a class but do not belong to that class. Errors of commission represent false positives.

Example

The code example below evaluates classifications using a confusion matrix.

PRO EvaluateClassificationUsingConfusionMatrix

    COMPILE_OPT IDL2

 

    ; Start the application

    e = ENVI()

 

    ; Open an input file

    File = Filepath('qb_boulder_msi', Subdir=['data'], $

    Root_Dir=e.Root_Dir)

    Raster = e.OpenRaster(File)

    File2 = Filepath('qb_boulder_roi.xml', Subdir=['data'], $

    Root_Dir=e.Root_Dir)

    Rois = envi.OpenROI(roiFile)

 

    ; Get training statistics

    StatTask = ENVITask('ROIStatistics')

    StatTask.INPUT_RASTER = Raster

    StatTask.INPUT_ROI = Rois

    StatTask.Execute

 

    ; Get the task from the catalog of ENVITasks

    Task = ENVITask('MahalanobisDistanceClassification')

 

    ; Define inputs

    Task.INPUT_RASTER = Raster

    Task.COVARIANCE = StatTask.Covariance

    Task.MEAN = StatTask.Mean

    Task.CLASS_PIXEL_COUNT = StatTask.Roi_Pixel_Count

    Task.CLASS_NAMES = [Rois[0].name, Rois[1].name, Rois[2].name]

    Task.CLASS_COLORS = [[0,0,255], [0,255,0], [255,0,0]]

 

    ; Run the task and display the result

    Task.Execute

    ClassRaster = Task.OUTPUT_RASTER

    View = e.GetView()

    Layer = View.CreateLayer(ClassRaster)

 

    ; Add the output to the Data Manager

    envi.Data.Add, ClassRaster

 

    ; Calculate the confusion matrix

    ConfusionMatrix = ENVICalculateConfusionMatrixFromRaster(ClassRaster, Rois)

 

    ; Print results

    Print, 'Confusion Matrix:'

    Print, ConfusionMatrix.Confusion_Matrix

    Print, 'Errors of commission: '

    Print, Transpose([[ConfusionMatrix.Column_Names+': '], [(ConfusionMatrix.CommissionError()).ToString()]])

    Print, 'Errors of omission: '

    Print, Transpose([[ConfusionMatrix.Column_Names+': '], [(ConfusionMatrix.OmissionError()).ToString()]])

    Print, 'Overall accuracy: ', ConfusionMatrix.Accuracy()

END

Syntax

Result = ENVIConfusionMatrix(Keywords=value)

Return Value

This function returns a reference to an ENVIConfusionMatrix object.

Arguments

None

Methods

Accuracy

ColumnTotals

CommissionError

F1

KappaCoefficient

OmissionError

Precision

ProducerAccuracy

Recall

RowTotals

UserAccuracy

Properties

COLUMN_NAMES (Get)

The truth class names.

CONFUSION_MATRIX (Get)

The confusion matrix computed from the input truth values and predicted values.

DESCRIPTION (Get, Set)

An optional description for the confusion matrix object.

ROW_NAMES (Get)

The predicted class names.

Keywords

The COLUMN_NAMES, DESCRIPTION, PREDICTED_VALUES, ROW_NAMES, and TRUTH_VALUES keywords are for users who want to manually specify confusion matrix parameters.

COLUMN_NAMES (optional)

Set this keyword to a string array of column names corresponding to the truth class names.

DESCRIPTION (optional)

Set this keyword to a string with a description for the confusion matrix.

ERROR (optional)

Set this keyword to a named variable that will contain any error message issued during execution of this routine. If no error occurs, the ERROR variable will be set to a null string (''). If an error occurs and the routine is a function, then the function result will be undefined.

When this keyword is not set and an error occurs, ENVI returns to the caller and execution halts. In this case, the error message is contained within !ERROR_STATE and can be caught using IDL's CATCH routine. See IDL Help for more information on !ERROR_STATE and CATCH.

See Manage Errors for more information on error handling in ENVI programming.

PREDICTED_VALUES (required)

Set this keyword to an array of predicted class values. The array size must be equal to that of TRUTH_VALUES.

ROW_NAMES (optional)

Set this keyword to a string array of row names corresponding to the predicted class names.

TRUTH_VALUES (required)

Set this keyword to an array of truth class values. The array size must be equal to that of PREDICTED_VALUES.

Version History

ENVI 5.4

Introduced

API Version

4.2