ClassificationClumping Task

This task performs a clumping method on a classification image. This operation clumps adjacent similar classified areas using morphological operators. Classified images often suffer from a lack of spatial coherency (speckle or holes in classified areas). Low pass filtering could be used to smooth these images, but the class information would be contaminated by adjacent class codes. Clumping classes solves this problem. The selected classes are clumped together by first performing a dilate operation then an erode operation on the classified image using one specified kernel (structuring element) for each operation.

Example

The following example performs an unsupervised classification, followed by a sieving, then clumping operation to remove the remaining black pixels.

; 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)

; Create a classification ENVIRaster

ClassTask = ENVITask('ISODATAClassification')

ClassTask.INPUT_RASTER = Raster

ClassTask.Execute

; Get the collection of data objects currently available in the Data Manager

DataColl = e.Data

; Add the class image to the Data Manager

DataColl.Add, ClassTask.OUTPUT_RASTER

; Display the result

View = e.GetView()

Layer = View.CreateLayer(ClassTask.OUTPUT_RASTER)

; Run the sieving task

SievingTask = ENVITask('ClassificationSieving')

SievingTask.INPUT_RASTER = ClassTask.OUTPUT_RASTER

SievingTask.Execute

; Run the clumping task

ClumpingTask = ENVITask('ClassificationClumping')

ClumpingTask.INPUT_RASTER = SievingTask.OUTPUT_RASTER

ClumpingTask.Execute

; Add the output to the Data Manager

DataColl.Add, ClumpingTask.OUTPUT_RASTER

; Display the result

Layer2 = View.CreateLayer(ClumpingTask.OUTPUT_RASTER)

; Flicker between the original classification and the result

; after clumping

Portal = View.CreatePortal()

Portal.Animate, 2.0, /FLICKER

Syntax

Result = ENVITask('ClassificationClumping')

Input properties (Set, Get): CLASS_ORDER, DILATE_KERNEL, ERODE_KERNEL, INPUT_RASTER, OUTPUT_RASTER_URI

Output properties (Get only):OUTPUT_RASTER

Properties marked as "Set" are those that you can set to specific values. You can also retrieve their current values any time. Properties marked as "Get" are those whose values you can retrieve but not set.

Methods

This task inherits the following methods from ENVITask:

Properties

This task inherits the following properties from ENVITask:

COMMUTE_ON_DOWNSAMPLE

CLASS_ORDER (optional)

Specify the order of class names in which clumping is applied to the classification image. If you do not specify this property, the classes are processed from first to last.

DILATE_KERNEL (required)

Specify a 2D array of zeros and ones that represents the structuring element (kernel) used for a dilate operation. If no kernel is specified, a 3 x 3 array will be used with a value of 1 for all of the array elements. Dilation is a morphological operation that uses a structuring element to expand the shapes contained in the input image.

ERODE_KERNEL (required)

Specify a 2D array of zeros and ones that represents the structuring element (kernel) used for a erode operation. If no kernel is specified, a 3 x 3 array will be used with a value of 1 for all of the array elements. Erosion is a morphological operation that uses a structuring element to reduce the shapes contained in the input image.

INPUT_RASTER (required)

Specify a raster on which to perform classification clumping.

OUTPUT_RASTER

This is a reference to the output raster of filetype ENVI.

OUTPUT_RASTER_URI (optional)

Specify a string with the fully qualified filename and path of the associated OUTPUT_RASTER. If you do not specify this property, or set it to an exclamation symbol (!), a temporary file will be created.

Version History

ENVI 5.2. 1

Introduced

API Version

4.2