de.unihalle.informatik.MiToBo.segmentation.snakes.optimize

Class SnakeOptimizerSingleVarCalc

java.lang.Object

de.unihalle.informatik.Alida.operator.ALDOperator

de.unihalle.informatik.Alida.operator.ALDOperatorControllable

de.unihalle.informatik.MiToBo.core.operator.MTBOperatorControllable

de.unihalle.informatik.MiToBo.segmentation.snakes.optimize.SnakeOptimizer

de.unihalle.informatik.MiToBo.segmentation.snakes.optimize.SnakeOptimizerSingle

de.unihalle.informatik.MiToBo.segmentation.snakes.optimize.SnakeOptimizerSingleVarCalc

All Implemented Interfaces:

de.unihalle.informatik.Alida.datatypes.ALDConfigurationValidator, de.unihalle.informatik.Alida.operator.events.ALDConfigurationEventListener, de.unihalle.informatik.Alida.operator.events.ALDConfigurationEventReporter, de.unihalle.informatik.Alida.operator.events.ALDControlEventListener, de.unihalle.informatik.Alida.operator.events.ALDControlEventReporter, de.unihalle.informatik.Alida.operator.events.ALDOperatorExecutionProgressEventListener, EventListener

@ALDAOperator(genericExecutionMode=ALL,
              level=STANDARD,
              shortDescription="Image contour segmentation with parametric snakes.")
 @ALDDerivedClass
public class SnakeOptimizerSingleVarCalc
extends SnakeOptimizerSingle

 
 Image contour segmentation using parametric snakes.
 
 This class provides methods to segment contours in an image based on
 parametric active contour models, i.e. snakes. Note that we assume here a
 counter-clockwise ordering of the snake points. Pay attention if you deal
 with upper-left coordinates!
 

 In this class snake optimization is done by solving PDEs like in the 
 original Kass et al. paper.
 
 TODO correct handling of not closed snakes
 TODO adaptation of gamma for every single point
 TODO adaptation of segment length
 
 

Author:

moeller, misiak

Nested Class Summary

Nested classes/interfaces inherited from class de.unihalle.informatik.MiToBo.segmentation.snakes.optimize.SnakeOptimizerSingle

SnakeOptimizerSingle.EnergyNormalizationMode, SnakeOptimizerSingle.IntensityNormalizationMode

Nested classes/interfaces inherited from class de.unihalle.informatik.MiToBo.segmentation.snakes.optimize.SnakeOptimizer

SnakeOptimizer.Snake_status

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

de.unihalle.informatik.Alida.operator.ALDOperatorControllable.OperatorControlStatus, de.unihalle.informatik.Alida.operator.ALDOperatorControllable.OperatorControlStatusHandle, de.unihalle.informatik.Alida.operator.ALDOperatorControllable.OperatorExecutionStatus

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 Jama.Matrix	A Optimization matrix A, to be modified by snake energies in each iteration.
private Jama.Matrix	B External energy vector, to be modified by snake energies.
protected MTBSet_SnakeEnergyDerivable	energySet List of snake energies to be used in segmentation.
protected double[]	energyWeightsNormed List of normalized energy weights.
protected double[][]	gammaAdaptive Vector containing the current gamma values (step sizes).
protected MTBGammaUpdate	gammaUpdater Gamma update strategy.
protected Double	initialGammas Initial step size in snake optimization.
private Jama.Matrix	memA Optimization matrix with values of last calculations.
protected MTBTermination	termCriterion Termination criterion.

Fields inherited from class de.unihalle.informatik.MiToBo.segmentation.snakes.optimize.SnakeOptimizerSingle

dispImg, doResampling, energy, energyCalculationRequested, intNormMode, normMode, previousEnergy, previousSnake, resampleSegLength, scaleFactor, snake, timer, workingImage

Fields inherited from class de.unihalle.informatik.MiToBo.segmentation.snakes.optimize.SnakeOptimizer

counterClockwiseSnakePointOrderRequested, energyData, excludeMask, iChannels, iHeight, inImg, initialSnakes, intermediateResults, itCounter, iWidth, outIntermediateResultsStack, outIntermediateResultsStackInterval, outIntermediateResultsStackWanted, outSnakes, outSnakesImg, sampleEnergyData, saveIntermediateResults, saveIntermediateResultsPath, showIntermediateResults, snakeNum

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

configurationEventlistenerList, controlEventlistenerList, notifyListenersRecursively, operatorExecStatus, stepSize, stepWiseExecution

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

completeDAG, name, operatorExecutionEventlistenerList, portHashAccess, verbose, versionProvider

Constructor Summary

Constructors

Constructor and Description
SnakeOptimizerSingleVarCalc() Default constructor.
SnakeOptimizerSingleVarCalc(MTBImage img, MTBPolygon2DSet initSnake, MTBSet_SnakeEnergyDerivable energs, MTBGammaUpdate gamUpdater, Double gammas, MTBTermination tc, Boolean resample, Double segLength) Default constructor with parameters.

Method Summary

Modifier and Type	Method and Description
protected double	calcSnakeEnergy() Calculates total snake energy.
SnakeOptimizerSingleVarCalc	clone()
protected SnakeOptimizer.Snake_status	doIteration() Performs one single optimization step.
double[][]	getCurGamma() Returns current gamma vector.
String	getDocumentation()
Vector<MTBSnakeEnergyDerivable>	getEnergies() Returns a reference to the list of energies.
MTBSet_SnakeEnergyDerivable	getEnergySet() Returns a reference to the set of energies.
Jama.Matrix	getMemorizedMatrixA() Returns matrix carrying always the values of the last iteration, or null before first run.
protected boolean	hasEnergies() Check if energies are given.
protected void	initOptimizer() Initializes the optimizer.
void	printParams() Print important parameters to standard output stream.
protected Object	readResolve() Function for proper initialization of deserialized objects.
void	resizeMatrices() Resize parameter matrices.
void	setEnergySet(MTBSet_ActiveContourEnergy eSet) Specify set of energies.
protected void	setupEnergyTable() Inits the table for sampled energy values.
String	toString()
void	validateCustom()

Methods inherited from class de.unihalle.informatik.MiToBo.segmentation.snakes.optimize.SnakeOptimizerSingle

closeWindows, disableEnergyCalculation, doResampleSnake, enableEnergyCalculation, getCurrentSnake, getCurrentSnakes, getEnergyValue, getIntensityNormalizationMode, getNormalizationMode, getPreviousEnergyValue, getPreviousSnake, getSegmentLength, getWorkingImage, normalizeInputImage, resampleSnake, saveSnake, setNormalizationMode, showSnake

Methods inherited from class de.unihalle.informatik.MiToBo.segmentation.snakes.optimize.SnakeOptimizer

disableSaveIntermediateResults, disableShowIntermediateResults, enableSaveIntermediateResults, enableShowIntermediateResults, getExcludeMask, getInitialSnakes, getInputImage, getIterationCount, getResultSnakeImage, getResultSnakes, getSnakeNumber, getStackWithIntermediateResults, operate, setExcludeMask, setInitialSnakes, setInputImage, setIntermediateResultPath, supportsStepWiseExecution, wantStackWithIntermediateResults

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

addALDConfigurationEventListener, addALDControlEventListener, fireALDConfigurationEvent, fireALDControlEvent, getControlStatus, getControlStatusHandle, getExecutionStatus, handleALDConfigurationEvent, handleALDControlEvent, removeALDConfigurationEventListener, removeALDControlEventListener, setControlStatus, setNotifyRecursiveFlag

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

addOperatorExecutionProgressEventListener, addParameter, addParameter, addParameterUnconditioned, fieldContained, fireOperatorExecutionProgressEvent, getALDPortHashAccessKey, getConstructionMode, 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, validateGeneric, writeHistory, writeHistory, writeHistory

Methods inherited from class java.lang.Object

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

Field Detail

energySet

@Parameter(label="List of Energies",
           mode=STANDARD,
           direction=IN,
           required=true,
           description="List of snake energies.",
           dataIOOrder=3)
protected MTBSet_SnakeEnergyDerivable energySet

List of snake energies to be used in segmentation.

gammaUpdater

@Parameter(label="Gamma Update Strategy",
           mode=ADVANCED,
           direction=IN,
           required=false,
           dataIOOrder=22,
           description="Gamma update object.")
protected MTBGammaUpdate gammaUpdater

Gamma update strategy.

termCriterion

@Parameter(label="Termination Criterion",
           mode=ADVANCED,
           direction=IN,
           required=false,
           dataIOOrder=20,
           description="Termination criterion for optimization.")
protected MTBTermination termCriterion

Termination criterion.

initialGammas

@Parameter(label="Initial Gamma Value",
           mode=STANDARD,
           direction=IN,
           required=false,
           dataIOOrder=21,
           description="Initial step sizes.")
protected Double initialGammas

Initial step size in snake optimization.

energyWeightsNormed

protected transient double[] energyWeightsNormed

List of normalized energy weights.

gammaAdaptive

protected transient double[][] gammaAdaptive

Vector containing the current gamma values (step sizes).

Note that the array is two-dimensional, however, the number of columns is always one! This is mainly for compatibility reasons as we need to convert the double array to Jama matrices which require two-dimensional arrays.

A

private transient Jama.Matrix A

Optimization matrix A, to be modified by snake energies in each iteration.

memA

private transient Jama.Matrix memA

Optimization matrix with values of last calculations.

B

private transient Jama.Matrix B

External energy vector, to be modified by snake energies.

Constructor Detail

SnakeOptimizerSingleVarCalc

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

Default constructor.

Throws:

de.unihalle.informatik.Alida.exceptions.ALDOperatorException

SnakeOptimizerSingleVarCalc

public SnakeOptimizerSingleVarCalc(MTBImage img,
                                   MTBPolygon2DSet initSnake,
                                   MTBSet_SnakeEnergyDerivable energs,
                                   MTBGammaUpdate gamUpdater,
                                   Double gammas,
                                   MTBTermination tc,
                                   Boolean resample,
                                   Double segLength)
                            throws de.unihalle.informatik.Alida.exceptions.ALDOperatorException

Default constructor with parameters.

Parameters:

img - Gray-scale input image.

initSnake - Array of initial snake polygon(s).

energs - Vector of energy objects.

gamUpdater - Object for gamma adaptation.

gammas - Initial value(s) for step size.

tc - Termination checker.

opStepWise - Enables step-through optimization (requires GUI).

stepsize - Number of iterations per step-through step.

normalize - Flag for enabling snake/image normalization.

resample - Flag for activating snake resampling.

segLength - Desired length of segments are resampling.

Throws:

de.unihalle.informatik.Alida.exceptions.ALDOperatorException

SnakeAlgosRegionException

Method Detail

clone

public SnakeOptimizerSingleVarCalc clone()

Specified by:

clone in class SnakeOptimizer

readResolve

protected Object readResolve()

Function for proper initialization of deserialized objects.

This function is called on an instance of this class being deserialized from file, prior to handing the instance over to the user. It takes care of a proper initialization of transient member variables as they are not initialized to the default values during deserialization.

Overrides:

readResolve in class SnakeOptimizerSingle

Returns:

validateCustom

public void validateCustom()
                    throws de.unihalle.informatik.Alida.exceptions.ALDOperatorException

Overrides:

validateCustom in class SnakeOptimizerSingle

Throws:

de.unihalle.informatik.Alida.exceptions.ALDOperatorException

hasEnergies

protected boolean hasEnergies()

Description copied from class: SnakeOptimizerSingle

Check if energies are given.

Specified by:

hasEnergies in class SnakeOptimizerSingle

Returns:

True, if energies are available.

printParams

public void printParams()

Print important parameters to standard output stream.

Overrides:

printParams in class SnakeOptimizerSingle

Throws:

de.unihalle.informatik.Alida.exceptions.ALDOperatorException

toString

public String toString()

Overrides:

toString in class SnakeOptimizerSingle

initOptimizer

protected void initOptimizer()
                      throws MTBSnakeException

Initializes the optimizer.

Here internal member variables are initialized according to the given parameters, and memory for intermediate results and debug data is allocated.

Overrides:

initOptimizer in class SnakeOptimizerSingle

Throws:

MTBSnakeException

doIteration

protected SnakeOptimizer.Snake_status doIteration()
                                           throws MTBException

Performs one single optimization step.

Specified by:

doIteration in class SnakeOptimizer

Returns:

Status of the optimization process.

Throws:

MTBException

resizeMatrices

public void resizeMatrices()

Resize parameter matrices.

The size of the matrices depends on the number of points used to represent the snake. This methods resizes the snake, if the given point number does not coincide with the current size of the matrices. After resizing the matrices, entries are set to zero.

setupEnergyTable

protected void setupEnergyTable()

Inits the table for sampled energy values.

calcSnakeEnergy

protected double calcSnakeEnergy()

Calculates total snake energy.

Returns:

Value of local energy, -1 if error during calculation.

setEnergySet

public void setEnergySet(MTBSet_ActiveContourEnergy eSet)

Description copied from class: SnakeOptimizerSingle

Specify set of energies.

Specified by:

setEnergySet in class SnakeOptimizerSingle

Parameters:

eSet - Set of energies.

getEnergySet

public MTBSet_SnakeEnergyDerivable getEnergySet()

Description copied from class: SnakeOptimizerSingle

Returns a reference to the set of energies.

Specified by:

getEnergySet in class SnakeOptimizerSingle

Returns:

Set of energies.

getEnergies

public Vector<MTBSnakeEnergyDerivable> getEnergies()

Returns a reference to the list of energies.

getCurGamma

public double[][] getCurGamma()

Returns current gamma vector.

getMemorizedMatrixA

public Jama.Matrix getMemorizedMatrixA()

Returns matrix carrying always the values of the last iteration, or null before first run.

getDocumentation

public String getDocumentation()

Overrides:

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