Package gov.nih.mipav.model.algorithms

Class AlgorithmVOISimplexOpt

java.lang.Object

java.lang.Thread

gov.nih.mipav.model.algorithms.AlgorithmBase

gov.nih.mipav.model.algorithms.AlgorithmVOISimplexOpt

All Implemented Interfaces:

de.jtem.numericalMethods.calculus.function.RealFunctionOfSeveralVariables, java.awt.event.ActionListener, java.awt.event.WindowListener, java.lang.Runnable, java.util.EventListener

public class AlgorithmVOISimplexOpt
extends AlgorithmBase
implements de.jtem.numericalMethods.calculus.function.RealFunctionOfSeveralVariables

Perform Simplex Optimization on a VOI that has been propagated to a slice to find a transformation which will make an affine "fit" a structure which the VOI fit on the previous slice (used in RFASegTool when propagating liver segmentation VOIs). Minimize cost = -sum of Gradient magnitude under propagated ROI in match image

Version:

1.0 July, 2003

Author:

Evan McCreedy

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
private class	AlgorithmVOISimplexOpt.CostFunction	Class for specifying optimization function for the simplex opt algorithm.

Nested classes/interfaces inherited from class java.lang.Thread

java.lang.Thread.State, java.lang.Thread.UncaughtExceptionHandler

Field Summary

Fields

Modifier and Type	Field	Description
protected float	baseVOIGradMagSum	The sum of the gradient magnitude along the polygon.
protected int	costFunc	Which cost function to use (MINDIFF or MAXSUM).
private int	currentSlice	The slice number the VOI is in.
private AlgorithmVOISimplexOpt.CostFunction	func
protected float[]	gradMagBuf	The gradient magnitude of the slice.
private ModelImage	image	The source image.
protected VOIContour	inputGon	The polygon of the VOI we are fitting.
static int	MAXSUM	Indicates that the maximum summation cost function should be used (this is the perferred cost function).
static int	MINDIFF	Indicates that the minimum difference cost function should be used (MAXSUM is perferred).
private float[]	sigmas	the kernel size to use in calculating the gradient magnitude image.
private int	simplexDim	Size of the simplex vertex array (5 == tx, ty, rot, zoomx, zoomy).
private float[]	sliceBuf	The image slice that holds the VOI we are working on.
protected int	xDim	The slice size in the x dimension.
protected int	yDim	The slice size in the y dimension.

Fields inherited from class gov.nih.mipav.model.algorithms.AlgorithmBase

destFlag, destImage, image25D, mask, maxProgressValue, minProgressValue, multiThreadingEnabled, nthreads, progress, progressModulus, progressStep, runningInSeparateThread, separable, srcImage, threadStopped

Fields inherited from class java.lang.Thread

MAX_PRIORITY, MIN_PRIORITY, NORM_PRIORITY

Constructor Summary

Constructors

Constructor	Description
AlgorithmVOISimplexOpt(ModelImage image, float[] sigmas, int costFunc)	Set up the algorithm to perform optimizations on an image.
AlgorithmVOISimplexOpt(ModelImage image, VOIContour inputGon, float[] sigmas, int costFunc)	Set up the algorithm to perform optimizations on an image.

Method Summary

Modifier and Type	Method	Description
double	eval(double[] x)
void	finalize()	Prepares this class for destruction.
int	getNumberOfVariables()
protected TransMatrix	getTransform(double[] x)	Converts a row of p into 3x3 transformation matrix.
VOIContour	goOptimize(float[] sliceBuf, VOIContour inputGon)	Run the simplex opt algorithm on a polygon (VOI) in a slice of an image.
void	runAlgorithm()	Run the algorithm on a polygon already specified through the constructor.
private VOIContour	search()	Find the best fit for the VOI polygon we have loaded into this algorithm.
void	setCurrentSlice(int currentSlice)	Specify the slice containing the VOI to optimize.
void	setPolygon(VOIContour poly)	Spefify the VOI polygon to optimize.

Methods inherited from class gov.nih.mipav.model.algorithms.AlgorithmBase

actionPerformed, addListener, addProgressChangeListener, calculateImageSize, calculatePrincipleAxis, computeElapsedTime, computeElapsedTime, convertIntoFloat, delinkProgressToAlgorithm, delinkProgressToAlgorithmMulti, displayError, errorCleanUp, fireProgressStateChanged, fireProgressStateChanged, fireProgressStateChanged, fireProgressStateChanged, fireProgressStateChanged, generateProgressValues, getDestImage, getElapsedTime, getMask, getMaxProgressValue, getMinProgressValue, getNumberOfThreads, getProgress, getProgressChangeListener, getProgressChangeListeners, getProgressModulus, getProgressStep, getProgressValues, getSrcImage, isCompleted, isImage25D, isMultiThreadingEnabled, isRunningInSeparateThread, isThreadStopped, linkProgressToAlgorithm, linkProgressToAlgorithm, makeProgress, notifyListeners, removeListener, removeProgressChangeListener, run, setCompleted, setImage25D, setMask, setMaxProgressValue, setMinProgressValue, setMultiThreadingEnabled, setNumberOfThreads, setProgress, setProgressModulus, setProgressStep, setProgressValues, setProgressValues, setRunningInSeparateThread, setSrcImage, setStartTime, setThreadStopped, startMethod, windowActivated, windowClosed, windowClosing, windowDeactivated, windowDeiconified, windowIconified, windowOpened

Methods inherited from class java.lang.Thread

activeCount, checkAccess, clone, countStackFrames, currentThread, dumpStack, enumerate, getAllStackTraces, getContextClassLoader, getDefaultUncaughtExceptionHandler, getId, getName, getPriority, getStackTrace, getState, getThreadGroup, getUncaughtExceptionHandler, holdsLock, interrupt, interrupted, isAlive, isDaemon, isInterrupted, join, join, join, onSpinWait, resume, setContextClassLoader, setDaemon, setDefaultUncaughtExceptionHandler, setName, setPriority, setUncaughtExceptionHandler, sleep, sleep, start, stop, suspend, toString, yield

Methods inherited from class java.lang.Object

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

Field Detail

MINDIFF

public static final int MINDIFF

Indicates that the minimum difference cost function should be used (MAXSUM is perferred).

See Also:

Constant Field Values

MAXSUM

public static final int MAXSUM

Indicates that the maximum summation cost function should be used (this is the perferred cost function).

See Also:

Constant Field Values

baseVOIGradMagSum

protected float baseVOIGradMagSum

The sum of the gradient magnitude along the polygon.

costFunc

protected int costFunc

Which cost function to use (MINDIFF or MAXSUM).

currentSlice

private int currentSlice

The slice number the VOI is in.

gradMagBuf

protected float[] gradMagBuf

The gradient magnitude of the slice.

image

private ModelImage image

The source image.

inputGon

protected VOIContour inputGon

The polygon of the VOI we are fitting.

sigmas

private float[] sigmas

the kernel size to use in calculating the gradient magnitude image.

simplexDim

private int simplexDim

Size of the simplex vertex array (5 == tx, ty, rot, zoomx, zoomy).

sliceBuf

private float[] sliceBuf

The image slice that holds the VOI we are working on.

xDim

protected int xDim

The slice size in the x dimension.

yDim

protected int yDim

The slice size in the y dimension.

func

private AlgorithmVOISimplexOpt.CostFunction func

Constructor Detail

AlgorithmVOISimplexOpt

public AlgorithmVOISimplexOpt(ModelImage image,
                              float[] sigmas,
                              int costFunc)

Set up the algorithm to perform optimizations on an image. The individual Polygons to optimize will be provided later (use goOptimize())

Parameters:

image - the image containing the VOI(s) to run the SimplexOpt algorithm on

sigmas - the kernel size to use in calculating the gradient magnitude image

costFunc - which cost function to use (MINDIFF or MAXSUM)

AlgorithmVOISimplexOpt

public AlgorithmVOISimplexOpt(ModelImage image,
                              VOIContour inputGon,
                              float[] sigmas,
                              int costFunc)

Set up the algorithm to perform optimizations on an image. To be used in conjunction with the run method.

Parameters:

image - the image containing the VOI(s) to run the SimplexOpt algorithm on

inputGon - the Polygon to optimize the fit of

sigmas - the kernel size to use in calculating the gradient magnitude image

costFunc - which cost function to use (MINDIFF or MAXSUM)

Method Detail

finalize

public void finalize()

Prepares this class for destruction.

Overrides:

finalize in class AlgorithmBase

goOptimize

public VOIContour goOptimize(float[] sliceBuf,
                             VOIContour inputGon)

Run the simplex opt algorithm on a polygon (VOI) in a slice of an image.

Parameters:

sliceBuf - the slice the polygon resides in

inputGon - the polygon to optimize the fit of

Returns:

the optimized polygon

runAlgorithm

public void runAlgorithm()

Run the algorithm on a polygon already specified through the constructor.

Specified by:

runAlgorithm in class AlgorithmBase

setCurrentSlice

public void setCurrentSlice(int currentSlice)

Specify the slice containing the VOI to optimize.

Parameters:

currentSlice - the slice with the VOI to optimize

setPolygon

public void setPolygon(VOIContour poly)

Spefify the VOI polygon to optimize.

Parameters:

poly - the polygon

getTransform

protected TransMatrix getTransform(double[] x)

Converts a row of p into 3x3 transformation matrix.

Parameters:

x - row of p[][] (one of the simplex vertices)

Returns:

the transformation matrix corresponding to x

search

private VOIContour search()

Find the best fit for the VOI polygon we have loaded into this algorithm.

Returns:

the optimized polygon

eval

public double eval(double[] x)

Specified by:

eval in interface de.jtem.numericalMethods.calculus.function.RealFunctionOfSeveralVariables

getNumberOfVariables

public int getNumberOfVariables()

Specified by:

getNumberOfVariables in interface de.jtem.numericalMethods.calculus.function.RealFunctionOfSeveralVariables