de.unihalle.informatik.MiToBo.fields

Class GradientFieldNonMaxSuppression

java.lang.Object

de.unihalle.informatik.Alida.operator.ALDOperator

de.unihalle.informatik.MiToBo.core.operator.MTBOperator

de.unihalle.informatik.MiToBo.fields.GradientFieldNonMaxSuppression

All Implemented Interfaces:

de.unihalle.informatik.Alida.datatypes.ALDConfigurationValidator, de.unihalle.informatik.Alida.operator.events.ALDOperatorExecutionProgressEventListener, EventListener

@ALDAOperator(genericExecutionMode=ALL,
              level=STANDARD,
              allowBatchMode=true)
public class GradientFieldNonMaxSuppression
extends MTBOperator

Non-Maximum-Suppression and (optional) hysteresis thresholding on gradient fields.

The input image is expected to include at least two channels, the first one with the components of the gradient field in x-direction and the second one with the components of the field in y-direction.

The operator first extracts pixel-wise gradient magnitudes and directions from the given input image channels and then removes all non-maximal gradients, e.g. sets the corresponding pixel positions in the result magnitude image to zero. Note that the gradient direction is discritized into intervals of 45 degrees for this procedure.

After non-maximum suppression an optional hysteresis thresholding may be applied which removes all pixels with gradient magnitudes below a lower threshold or lying between the upper and lower threshold but not being linked by a path to a pixel exceeding the upper threshold.

As result a gradient magnitude image is returned which only contains values different from zero at positions with locally maximal gradients and which survived the (optional) hysteresis thresholding step.

Author:

Birgit Moeller

Nested Class Summary

Nested classes/interfaces inherited from class de.unihalle.informatik.Alida.operator.ALDOperator

de.unihalle.informatik.Alida.operator.ALDOperator.HidingMode, de.unihalle.informatik.Alida.operator.ALDOperator.HistoryConstructionMode

Field Summary

Fields

Modifier and Type	Field and Description
private boolean	doHysteresisThresholding Flag to enable/disable additional hysteresis thresholding.
private MTBImageDouble	resultImage Resulting thinned (and thresholded) magnitude image.
private MTBImage	vectorFieldImage Image with input 2D gradient field.

Fields inherited from class de.unihalle.informatik.Alida.operator.ALDOperator

completeDAG, name, operatorExecutionEventlistenerList, portHashAccess, verbose, versionProvider

Constructor Summary

Constructors

Constructor and Description
GradientFieldNonMaxSuppression() Standard constructor.

Method Summary

Modifier and Type	Method and Description
protected MTBImageDouble	doHS(MTBImageDouble magImg) Does the hysteresis thresholding.
protected MTBImageDouble	doNMS() Applies non-maximum-suppression to the input data.
MTBImageDouble	getResultImage() Returns the result image.
protected static void	labelNeighbors(MTBImage magImg, MTBImage hsImg, int x, int y, double lt, double ht) Function for recursive labeling of contour pixels.
protected void	operate()

Methods inherited from class de.unihalle.informatik.MiToBo.core.operator.MTBOperator

readResolve

Methods inherited from class de.unihalle.informatik.Alida.operator.ALDOperator

addOperatorExecutionProgressEventListener, addParameter, addParameter, addParameterUnconditioned, fieldContained, fireOperatorExecutionProgressEvent, getALDPortHashAccessKey, getConstructionMode, getDocumentation, getHidingMode, getInactiveParameterNames, getInInoutNames, getInInoutNames, getInNames, getInOutNames, getMissingRequiredInputs, getName, getNumParameters, getOutInoutNames, getOutNames, getParameter, getParameterDescriptor, getParameterDescriptorUnconditioned, getParameterNames, getParameterUnconditioned, getSupplementalNames, getVerbose, getVersion, handleOperatorExecutionProgressEvent, hasInOutParameters, hasParameter, isAnnotatedParameter, isConfigured, print, print, print, printInterface, printInterface, readHistory, reinitializeParameterDescriptors, removeOperatorExecutionProgressEventListener, removeParameter, runOp, runOp, runOp, setConstructionMode, setConstructionMode, setConstructionMode, setHidingMode, setName, setParameter, setParameterUnconditioned, setVerbose, toStringVerbose, unconfiguredItems, validate, validateCustom, validateGeneric, writeHistory, writeHistory, writeHistory

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

vectorFieldImage

@Parameter(label="Gradient Vector Field Image",
           required=true,
           direction=IN,
           dataIOOrder=0,
           description="Gradient field image.")
private transient MTBImage vectorFieldImage

Image with input 2D gradient field.

doHysteresisThresholding

@Parameter(label="Do hysteresis thresholding?",
           required=true,
           direction=IN,
           dataIOOrder=1,
           description="Enable/disable additional hysteresis thresholding.")
private boolean doHysteresisThresholding

Flag to enable/disable additional hysteresis thresholding.

resultImage

@Parameter(label="Thinned and thresholded magnitude image.",
           direction=OUT,
           description="Resulting magnitude image")
private transient MTBImageDouble resultImage

Resulting thinned (and thresholded) magnitude image.

Constructor Detail

GradientFieldNonMaxSuppression

public GradientFieldNonMaxSuppression()
                               throws de.unihalle.informatik.Alida.exceptions.ALDOperatorException

Standard constructor. A new empty operator object is initialized.

Throws:

de.unihalle.informatik.Alida.exceptions.ALDOperatorException - Thrown in case of failure.

Method Detail

operate

protected void operate()

Specified by:

operate in class de.unihalle.informatik.Alida.operator.ALDOperator

doNMS

protected MTBImageDouble doNMS()

Applies non-maximum-suppression to the input data.

Returns:

Filtered image with gradient magnitudes.

doHS

protected MTBImageDouble doHS(MTBImageDouble magImg)

Does the hysteresis thresholding.

Parameters:

magImg - Image with gradient magnitudes.

Returns:

Result image.

labelNeighbors

protected static void labelNeighbors(MTBImage magImg,
                                     MTBImage hsImg,
                                     int x,
                                     int y,
                                     double lt,
                                     double ht)

Function for recursive labeling of contour pixels.

Refer to the hysteresis threshold algorithm for details.

Parameters:

magImg - Image with gradient magnitudes.

hsImg - Current hysteresis filtered image.

x - Current x-position.

y - Current y-position.

lt - Lower threshold.

ht - Higher threshold.

getResultImage

public MTBImageDouble getResultImage()

Returns the result image.

Returns:

Result image with thinned gradient magnitudes.