Package gov.nih.mipav.model.algorithms

Class AlgorithmEntropyMinimization

java.lang.Object

java.lang.Thread

gov.nih.mipav.model.algorithms.AlgorithmBase

gov.nih.mipav.model.algorithms.AlgorithmEntropyMinimization

All Implemented Interfaces:

de.jtem.numericalMethods.calculus.function.RealFunctionOfSeveralVariables, ActionListener, WindowListener, Runnable, EventListener

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

DOCUMENT ME!

Nested Class Summary

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

Thread.Builder, Thread.State, Thread.UncaughtExceptionHandler

Field Summary

Fields

Modifier and Type	Field	Description
private double	ac1	DOCUMENT ME!
private double	ac2	DOCUMENT ME!
private double	ac3	DOCUMENT ME!
private double	ac4	DOCUMENT ME!
private double	ac5	DOCUMENT ME!
private double	ac6	DOCUMENT ME!
private double	ac7	DOCUMENT ME!
private double	ac8	DOCUMENT ME!
private double	ac9	DOCUMENT ME!
private double	ad1	DOCUMENT ME!
private double	ad2	DOCUMENT ME!
private double	ad3	DOCUMENT ME!
private double	ad4	DOCUMENT ME!
private double	ad5	DOCUMENT ME!
private double	ad6	DOCUMENT ME!
private double	ad7	DOCUMENT ME!
private double	ad8	DOCUMENT ME!
private double	ad9	DOCUMENT ME!
private int	area	DOCUMENT ME!
private boolean	backgroundPresent	DOCUMENT ME!
private float[]	buffer	DOCUMENT ME!
private double	grayCount	DOCUMENT ME!
private int	histogramAvailable	Number of pixels available for histogram formation.
private double[]	idealBuffer	Buffer holding calculation for noiseless U(x,y).
private int	iter	DOCUMENT ME!
private double	maximum	original image maximum.
private double	maximumB	DOCUMENT ME!
private double	maximumG	DOCUMENT ME!
private double	maximumR	DOCUMENT ME!
private double	mc1	DOCUMENT ME!
private double	mc10	DOCUMENT ME!
private double	mc11	DOCUMENT ME!
private double	mc12	DOCUMENT ME!
private double	mc13	DOCUMENT ME!
private double	mc14	DOCUMENT ME!
private double	mc15	DOCUMENT ME!
private double	mc16	DOCUMENT ME!
private double	mc17	DOCUMENT ME!
private double	mc18	DOCUMENT ME!
private double	mc19	DOCUMENT ME!
private double	mc2	DOCUMENT ME!
private double	mc3	DOCUMENT ME!
private double	mc4	DOCUMENT ME!
private double	mc5	DOCUMENT ME!
private double	mc6	DOCUMENT ME!
private double	mc7	DOCUMENT ME!
private double	mc8	DOCUMENT ME!
private double	mc9	DOCUMENT ME!
private double	md1	DOCUMENT ME!
private double	md10	DOCUMENT ME!
private double	md11	DOCUMENT ME!
private double	md12	DOCUMENT ME!
private double	md13	DOCUMENT ME!
private double	md14	DOCUMENT ME!
private double	md15	DOCUMENT ME!
private double	md16	DOCUMENT ME!
private double	md17	DOCUMENT ME!
private double	md18	DOCUMENT ME!
private double	md19	DOCUMENT ME!
private double	md2	DOCUMENT ME!
private double	md3	DOCUMENT ME!
private double	md4	DOCUMENT ME!
private double	md5	DOCUMENT ME!
private double	md6	DOCUMENT ME!
private double	md7	DOCUMENT ME!
private double	md8	DOCUMENT ME!
private double	md9	DOCUMENT ME!
private double	minimum	original image minimum.
private double	minimumB	DOCUMENT ME!
private double	minimumG	DOCUMENT ME!
private double	minimumR	DOCUMENT ME!
static final int	NOISE_M2	Quadratic multiplicative noise.
static final int	NOISE_M3	Cubic multiplicative noise.
static final int	NOISE_MA2	The user can choose to use thresholding to exclude the background from any processing.
private int	noiseType	NOISE_MA2 or NOISE_M2.
private int	nParams	Numbering of shading correction parameters.
private boolean[]	objectBuffer	DOCUMENT ME!
private int	pixelIncrement	pixelIncrement is increased above 1 if subsampling.
private double	powellTolerance	Tolerance passed to Powell's algorithm.
private int	sliceSize	xDim * yDim.
private boolean	subsample	If true, subsample image for histogram formation.
private float	thresholdLevel	DOCUMENT ME!
private boolean	thresholdSelected	If true don't use pixels below thresholdLevel that are 4 or 6 neighbor connected to boundary pixels below thresholdLevel.
private boolean[]	validBuffer	True for source input locations not having image min or image max values.
private int	volSize	DOCUMENT ME!
private int	volume	DOCUMENT ME!
private int	xDim	DOCUMENT ME!
private int	yDim	DOCUMENT ME!
private int	zDim	DOCUMENT ME!
private int	zPixelIncrement	Handle case where zDim is far less than xDim and yDim

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
AlgorithmEntropyMinimization(ModelImage srcImage, boolean thresholdSelected, float thresholdLevel, boolean subsample, int noiseType)	Creates a new AlgorithmEntropyMinimization object.
AlgorithmEntropyMinimization(ModelImage resultImage, ModelImage srcImage, boolean thresholdSelected, float thresholdLevel, boolean subsample, int noiseType)	Creates a new AlgorithmEntropyMinimization object.

Method Summary

Modifier and Type	Method	Description
private double	entropyFunction(double[] p)	DOCUMENT ME!
double	eval(double[] x)
void	finalize()	Prepares this class for destruction.
int	getNumberOfVariables()
private void	powell(double[] p, double[][] xi, double ftol)	Minimization of entropy function of nParams variables.
private void	run2Da2m2()	DOCUMENT ME!
private void	run2Dm2()	DOCUMENT ME!
private void	run2Dm3()	DOCUMENT ME!
private void	run3Da2m2()	DOCUMENT ME!
private void	run3Dm2()	DOCUMENT ME!
private void	run3Dm3()	DOCUMENT ME!
void	runAlgorithm()	DOCUMENT ME!

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, isVirtual, join, join, join, join, ofPlatform, ofVirtual, onSpinWait, resume, setContextClassLoader, setDaemon, setDefaultUncaughtExceptionHandler, setName, setPriority, setUncaughtExceptionHandler, sleep, sleep, sleep, start, startVirtualThread, stop, suspend, threadId, toString, yield

Methods inherited from class java.lang.Object

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

Field Details

NOISE_MA2

public static final int NOISE_MA2

The user can choose to use thresholding to exclude the background from any processing. Excluded pixels will have the same values as the original source image. All pixels with values below threshold that are 4 or 6 neighbor connected thru below threshold pixels to a boundary pixel which is below threshold will be excluded from processing. By default image subsampling is used in histogram formation since using 5000 images samples is sufficient to form the histogram. If subsampling is selected, pixelIncrement for both x and y is set equal to: Math.max(1,(int)Math.sqrt(histogramAvailable/5000)) If subsampling is not selected, pixelIncrement equals 1 Powell's algorithm is used to search for the values of the 10 shading parameters that minimize the entropy of the blurred histogram of the U(x,y) image. A histogram is formed to represent integers from the floor(image U minimum) to the ceiling(image U maximum). For each floating point gray value gp in U(x,y) a count of 1 is split between the 2 integer bins encompassing it. Let the bins be k and k+1. k+1-gp is added to bin k and gp - k is added to bin k+1. Gray values in locations corresponding to the image minimum or image maximum in the original input image are not used in creating this histogram, since they may represent regions where saturation effects are occurring. Then, a blurred histogram is created from the original histogram. The original histogram is blurred with a triangular window of size 5: blurredHistogram[i] = originalHistogram[i-2] + 2* originalHistogram[i-1] + 3*originalHistogram[i] + 2*originalHistogram[i+1] + originalHistogram[i+2] The blurred histogram is normalized to give probabilities by dividing the count in each bin by the total number of counts in the blurred histogram. The entropy is formed by summing the -probability(n)*log(probability(n)) over all bins n with nonzero counts. References: This software implements the algorithm discussed in: 1.) Retrospective shading correction based on entropy minimization by B. Likar, J. B. A. Maintz, M. A. Viergever, and F. Pernus, Journal of Microscopy, Vol. 197, Pt 3., March, 2000, pp. 285-295. 2.) Retrospective Correction of MR Intensity Inhomogeneity by Information Minimization by Bostjan Likar, Max Viergever, and Franjo Pernus, IEEE Transactions on Medical Imaging, Vol. 20, No. 12, December, 2001, pp. 1398-1410. 3.) Real-time automated visual inspection of color tablets in pharmaceutical blisters by Joze Derganc, Bostjan Likar, Rok Bernard, Dejan Tomazevic, and Franjo Pernus, Real-Time Imaging, Vol. 9, 2003, pp. 113-124.

See Also:

• Constant Field Values

NOISE_M2

public static final int NOISE_M2

Quadratic multiplicative noise.

See Also:

• Constant Field Values

NOISE_M3

public static final int NOISE_M3

Cubic multiplicative noise.

See Also:

• Constant Field Values

ac1

private double ac1

DOCUMENT ME!

ac2

private double ac2

DOCUMENT ME!

ac3

private double ac3

DOCUMENT ME!

ac4

private double ac4

DOCUMENT ME!

ac5

private double ac5

DOCUMENT ME!

ac6

private double ac6

DOCUMENT ME!

ac7

private double ac7

DOCUMENT ME!

ac8

private double ac8

DOCUMENT ME!

ac9

private double ac9

DOCUMENT ME!

ad1

private double ad1

DOCUMENT ME!

ad2

private double ad2

DOCUMENT ME!

ad3

private double ad3

DOCUMENT ME!

ad4

private double ad4

DOCUMENT ME!

ad5

private double ad5

DOCUMENT ME!

ad6

private double ad6

DOCUMENT ME!

ad7

private double ad7

DOCUMENT ME!

ad8

private double ad8

DOCUMENT ME!

ad9

private double ad9

DOCUMENT ME!

area

private int area

DOCUMENT ME!

backgroundPresent

private boolean backgroundPresent

DOCUMENT ME!

buffer

private float[] buffer

DOCUMENT ME!

grayCount

private double grayCount

DOCUMENT ME!

histogramAvailable

private int histogramAvailable

Number of pixels available for histogram formation.

idealBuffer

private double[] idealBuffer

Buffer holding calculation for noiseless U(x,y).

iter

private int iter

DOCUMENT ME!

maximum

private double maximum

original image maximum.

maximumR

private double maximumR

DOCUMENT ME!

maximumG

private double maximumG

DOCUMENT ME!

maximumB

private double maximumB

DOCUMENT ME!

mc1

private double mc1

DOCUMENT ME!

mc2

private double mc2

DOCUMENT ME!

mc3

private double mc3

DOCUMENT ME!

mc4

private double mc4

DOCUMENT ME!

mc5

private double mc5

DOCUMENT ME!

mc6

private double mc6

DOCUMENT ME!

mc7

private double mc7

DOCUMENT ME!

mc8

private double mc8

DOCUMENT ME!

mc9

private double mc9

DOCUMENT ME!

mc10

private double mc10

DOCUMENT ME!

mc11

private double mc11

DOCUMENT ME!

mc12

private double mc12

DOCUMENT ME!

mc13

private double mc13

DOCUMENT ME!

mc14

private double mc14

DOCUMENT ME!

mc15

private double mc15

DOCUMENT ME!

mc16

private double mc16

DOCUMENT ME!

mc17

private double mc17

DOCUMENT ME!

mc18

private double mc18

DOCUMENT ME!

mc19

private double mc19

DOCUMENT ME!

md1

private double md1

DOCUMENT ME!

md2

private double md2

DOCUMENT ME!

md3

private double md3

DOCUMENT ME!

md4

private double md4

DOCUMENT ME!

md5

private double md5

DOCUMENT ME!

md6

private double md6

DOCUMENT ME!

md7

private double md7

DOCUMENT ME!

md8

private double md8

DOCUMENT ME!

md9

private double md9

DOCUMENT ME!

md10

private double md10

DOCUMENT ME!

md11

private double md11

DOCUMENT ME!

md12

private double md12

DOCUMENT ME!

md13

private double md13

DOCUMENT ME!

md14

private double md14

DOCUMENT ME!

md15

private double md15

DOCUMENT ME!

md16

private double md16

DOCUMENT ME!

md17

private double md17

DOCUMENT ME!

md18

private double md18

DOCUMENT ME!

md19

private double md19

DOCUMENT ME!

minimum

private double minimum

original image minimum.

minimumR

private double minimumR

DOCUMENT ME!

minimumG

private double minimumG

DOCUMENT ME!

minimumB

private double minimumB

DOCUMENT ME!

noiseType

private int noiseType

NOISE_MA2 or NOISE_M2.

nParams

private int nParams

Numbering of shading correction parameters.

objectBuffer

private boolean[] objectBuffer

DOCUMENT ME!

pixelIncrement

private int pixelIncrement

pixelIncrement is increased above 1 if subsampling.

zPixelIncrement

private int zPixelIncrement

Handle case where zDim is far less than xDim and yDim

powellTolerance

private double powellTolerance

Tolerance passed to Powell's algorithm.

sliceSize

private int sliceSize

xDim * yDim.

subsample

private boolean subsample

If true, subsample image for histogram formation.

thresholdLevel

private float thresholdLevel

DOCUMENT ME!

thresholdSelected

private boolean thresholdSelected

If true don't use pixels below thresholdLevel that are 4 or 6 neighbor connected to boundary pixels below thresholdLevel.

validBuffer

private boolean[] validBuffer

True for source input locations not having image min or image max values.

volSize

private int volSize

DOCUMENT ME!

volume

private int volume

DOCUMENT ME!

xDim

private int xDim

DOCUMENT ME!

yDim

private int yDim

DOCUMENT ME!

zDim

private int zDim

DOCUMENT ME!

Constructor Details

AlgorithmEntropyMinimization

public AlgorithmEntropyMinimization(ModelImage srcImage,
 boolean thresholdSelected,
 float thresholdLevel,
 boolean subsample,
 int noiseType)

Creates a new AlgorithmEntropyMinimization object.

Parameters:

srcImage - original image

thresholdSelected - If true, don't use pixels that are 4 or 6 neighbor connected to boundary pixels below thresholdLevel

thresholdLevel - DOCUMENT ME!

subsample - DOCUMENT ME!

noiseType - DOCUMENT ME!

AlgorithmEntropyMinimization

public AlgorithmEntropyMinimization(ModelImage resultImage,
 ModelImage srcImage,
 boolean thresholdSelected,
 float thresholdLevel,
 boolean subsample,
 int noiseType)

Creates a new AlgorithmEntropyMinimization object.

Parameters:

resultImage - entropy minimized image

srcImage - original image

thresholdSelected - If true, don't use pixels that are 4 or 6 neighbor connected to boundary pixels below thresholdLevel

thresholdLevel - DOCUMENT ME!

subsample - DOCUMENT ME!

noiseType - DOCUMENT ME!

Method Details

finalize

public void finalize()

Prepares this class for destruction.

Overrides:

finalize in class AlgorithmBase

runAlgorithm

public void runAlgorithm()

DOCUMENT ME!

Specified by:

runAlgorithm in class AlgorithmBase

entropyFunction

private double entropyFunction(double[] p)

DOCUMENT ME!

Parameters:

p - DOCUMENT ME!

Returns:

DOCUMENT ME!

powell

private void powell(double[] p,
 double[][] xi,
 double ftol)

Minimization of entropy function of nParams variables. Input consists of an initial starting point p[0...nParams-1]; an initial matrix xi[0...nParams-1][0...nParams-1], whose columns contain the initial set of directions (usually the nParams unit vectors); and ftol, the fractional tolerance in the function value such that failure to decrease by more than this amount on one iteration signals doneness. On output, p is set to the best point found, xi is the then-current direction set, fret is the returned function value at p, and iter is the number of iterations taken.

Parameters:

p - DOCUMENT ME!

xi - DOCUMENT ME!

ftol - DOCUMENT ME!

run2Da2m2

private void run2Da2m2()

DOCUMENT ME!

run2Dm2

private void run2Dm2()

DOCUMENT ME!

run2Dm3

private void run2Dm3()

DOCUMENT ME!

run3Da2m2

private void run3Da2m2()

DOCUMENT ME!

run3Dm2

private void run3Dm2()

DOCUMENT ME!

run3Dm3

private void run3Dm3()

DOCUMENT ME!

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