mpicbg.imglib.algorithm.scalespace

Class DifferenceOfGaussian<A extends Type<A>,B extends NumericType<B> & Comparable<B>>

java.lang.Object

mpicbg.imglib.algorithm.scalespace.DifferenceOfGaussian<A,B>

All Implemented Interfaces:

Algorithm, Benchmark, MultiThreaded

Direct Known Subclasses:

DifferenceOfGaussianReal

public class DifferenceOfGaussian<A extends Type<A>,B extends NumericType<B> & Comparable<B>>
extends Object
implements Algorithm, MultiThreaded, Benchmark

TODO

Author:

Stephan Preibisch

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	DifferenceOfGaussian.SpecialPoint

Field Summary

Fields

Modifier and Type	Field and Description
protected Converter<A,B>	converter
protected Image<B>	dogImage
protected ImageFactory<B>	factory
protected Image<A>	image
protected OutOfBoundsStrategyFactory<B>	outOfBoundsFactory
protected ArrayList<DifferenceOfGaussianPeak<B>>	peaks

Constructor Summary

Constructors

Constructor and Description
DifferenceOfGaussian(Image<A> img, ImageFactory<B> factory, Converter<A,B> converter, OutOfBoundsStrategyFactory<B> outOfBoundsFactory, double[] sigma1, double[] sigma2, B minPeakValue, B normalizationFactor) Extracts local minima and maxima of a certain size.
DifferenceOfGaussian(Image<A> img, ImageFactory<B> factory, Converter<A,B> converter, OutOfBoundsStrategyFactory<B> outOfBoundsFactory, double sigma1, double sigma2, B minPeakValue, B normalizationFactor) Calls the DifferenceOfGaussian constructor with the given sigmas copied into double[] arrays, one entry per dimension.

Method Summary

Modifier and Type	Method and Description
boolean	checkInput()
ArrayList<DifferenceOfGaussianPeak<B>>	findPeaks(Image<B> laPlace)
boolean	getComputeConvolutionsParalell()
Image<B>	getDoGImage()
String	getErrorMessage()
protected OutputAlgorithm<B>	getGaussianConvolution(double[] sigma, int nThreads) This method returns the OutputAlgorithm that will compute the Gaussian Convolutions, more efficient versions can override this method
boolean	getKeepDoGImage()
B	getMinPeakValue()
protected Function<B,B,B>	getNormalizedSubtraction() Returns the function that does the normalized subtraction of the gauss images, more efficient versions can override this method
int	getNumThreads() The number of threads used by the algorithm
ArrayList<DifferenceOfGaussianPeak<B>>	getPeaks()
long	getProcessingTime()
protected boolean	isPeakHighEnough(B value) Checks if the absolute value of the current peak is high enough, more efficient versions can override this method
protected DifferenceOfGaussian.SpecialPoint	isSpecialPoint(LocalNeighborhoodCursor<B> neighborhoodCursor, B centerValue) Checks if the current position is a minima or maxima in a 3^n neighborhood, more efficient versions can override this method
boolean	process()
void	setComputeConvolutionsParalell(boolean paralell)
void	setKeepDoGImage(boolean keepDoGImage)
void	setMinPeakValue(B value)
void	setNumThreads() Sets the number of threads to the amount of processors available
void	setNumThreads(int numThreads) Sets the number of threads

Methods inherited from class java.lang.Object

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

Field Detail

image

protected final Image<A extends Type<A>> image

dogImage

protected Image<B extends NumericType<B> & Comparable<B>> dogImage

factory

protected final ImageFactory<B extends NumericType<B> & Comparable<B>> factory

outOfBoundsFactory

protected final OutOfBoundsStrategyFactory<B extends NumericType<B> & Comparable<B>> outOfBoundsFactory

peaks

protected final ArrayList<DifferenceOfGaussianPeak<B extends NumericType<B> & Comparable<B>>> peaks

converter

protected final Converter<A extends Type<A>,B extends NumericType<B> & Comparable<B>> converter

Constructor Detail

DifferenceOfGaussian

public DifferenceOfGaussian(Image<A> img,
                            ImageFactory<B> factory,
                            Converter<A,B> converter,
                            OutOfBoundsStrategyFactory<B> outOfBoundsFactory,
                            double sigma1,
                            double sigma2,
                            B minPeakValue,
                            B normalizationFactor)

Calls the DifferenceOfGaussian constructor with the given sigmas copied into double[] arrays, one entry per dimension.

DifferenceOfGaussian

public DifferenceOfGaussian(Image<A> img,
                            ImageFactory<B> factory,
                            Converter<A,B> converter,
                            OutOfBoundsStrategyFactory<B> outOfBoundsFactory,
                            double[] sigma1,
                            double[] sigma2,
                            B minPeakValue,
                            B normalizationFactor)

Extracts local minima and maxima of a certain size. It therefore computes the difference of gaussian for an Image and detects all local minima and maxima in 3x3x3x....3 environment, which is returned as an ArrayList of DifferenceOfGaussianPeaks. The two sigmas define the scale on which extrema are identified, it correlates with the size of the object. Note that not only extrema of this size are found, but they will have the higher absolute values. Note as well that the values of the difference of gaussian image is also defined by the distance between the two sigmas. A normalization if necessary can be found in the ScaleSpace class. Also note a technical detail, the method findPeaks(Image img) can be called on any image if the image from where the extrema should be computed already exists.

Parameters:

img - - The input Image

factory - - The ImageFactory which defines the datatype in which the computation is performed

converter - - The Converter which defines how to convert into

outOfBoundsFactory - - The OutOfBoundsStrategyFactory necessary for the GaussianConvolution

sigma1 - - The lower sigma

sigma2 - - The higher sigma

minPeakValue - -

normalizationFactor -

Method Detail

setMinPeakValue

public void setMinPeakValue(B value)

getMinPeakValue

public B getMinPeakValue()

getDoGImage

public Image<B> getDoGImage()

setKeepDoGImage

public void setKeepDoGImage(boolean keepDoGImage)

getKeepDoGImage

public boolean getKeepDoGImage()

getPeaks

public ArrayList<DifferenceOfGaussianPeak<B>> getPeaks()

setComputeConvolutionsParalell

public void setComputeConvolutionsParalell(boolean paralell)

getComputeConvolutionsParalell

public boolean getComputeConvolutionsParalell()

getGaussianConvolution

protected OutputAlgorithm<B> getGaussianConvolution(double[] sigma,
                                                    int nThreads)

This method returns the OutputAlgorithm that will compute the Gaussian Convolutions, more efficient versions can override this method

Parameters:

sigma - - the sigma of the convolution

nThreads - - the number of threads for this convolution

Returns:

getNormalizedSubtraction

protected Function<B,B,B> getNormalizedSubtraction()

Returns the function that does the normalized subtraction of the gauss images, more efficient versions can override this method

Returns:

- the Subtraction Function

isPeakHighEnough

protected boolean isPeakHighEnough(B value)

Checks if the absolute value of the current peak is high enough, more efficient versions can override this method

Parameters:

value - - the current value

Returns:

true if the absoluted value is high enough, otherwise false

isSpecialPoint

protected DifferenceOfGaussian.SpecialPoint isSpecialPoint(LocalNeighborhoodCursor<B> neighborhoodCursor,
                                                           B centerValue)

Checks if the current position is a minima or maxima in a 3^n neighborhood, more efficient versions can override this method

Parameters:

neighborhoodCursor - - the LocalNeighborhoodCursor

centerValue - - the value in the center which is tested

Returns:

- if is a minimum, maximum or nothig

process

public boolean process()

Specified by:

process in interface Algorithm

findPeaks

public ArrayList<DifferenceOfGaussianPeak<B>> findPeaks(Image<B> laPlace)

checkInput

public boolean checkInput()

Specified by:

checkInput in interface Algorithm

getErrorMessage

public String getErrorMessage()

Specified by:

getErrorMessage in interface Algorithm

getProcessingTime

public long getProcessingTime()

Specified by:

getProcessingTime in interface Benchmark

setNumThreads

public void setNumThreads()

Description copied from interface: MultiThreaded

Sets the number of threads to the amount of processors available

Specified by:

setNumThreads in interface MultiThreaded

setNumThreads

public void setNumThreads(int numThreads)

Description copied from interface: MultiThreaded

Sets the number of threads

Specified by:

setNumThreads in interface MultiThreaded

Parameters:

numThreads - - number of threads to use

getNumThreads

public int getNumThreads()

Description copied from interface: MultiThreaded

The number of threads used by the algorithm

Specified by:

getNumThreads in interface MultiThreaded

Returns:

- the number of threads