cern.colt.matrix.tfloat.impl

Class DenseLargeFloatMatrix3D

java.lang.Object

cern.colt.PersistentObject

cern.colt.matrix.AbstractMatrix

cern.colt.matrix.AbstractMatrix3D

cern.colt.matrix.tfloat.FloatMatrix3D

cern.colt.matrix.tfloat.impl.WrapperFloatMatrix3D

cern.colt.matrix.tfloat.impl.DenseLargeFloatMatrix3D

All Implemented Interfaces:

Serializable, Cloneable

public class DenseLargeFloatMatrix3D
extends WrapperFloatMatrix3D

Dense 3-d matrix holding float elements. First see the package summary and javadoc tree view to get the broad picture.

Implementation:

This data structure allows to store more than 2^31 elements. Internally holds one three-dimensional array, elements[slices][rows][columns]. Note that this implementation is not synchronized.

Time complexity:

O(1) (i.e. constant time) for the basic operations get, getQuick, set, setQuick and size.

Author:

Piotr Wendykier (piotr.wendykier@gmail.com)

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
DenseLargeFloatMatrix3D(int slices, int rows, int columns)

Method Summary

Methods

Modifier and Type	Method and Description
void	dct2Slices(boolean scale) Computes the 2D discrete cosine transform (DCT-II) of each slice of this matrix.
void	dct3(boolean scale) Computes the 3D discrete cosine transform (DCT-II) of this matrix.
void	dht2Slices() Computes the 2D discrete Hartley transform (DHT) of each slice of this matrix.
void	dht3() Computes the 3D discrete Hartley transform (DHT) of this matrix.
void	dst2Slices(boolean scale) Computes the 2D discrete sine transform (DST-II) of each slice of this matrix.
void	dst3(boolean scale) Computes the 3D discrete sine transform (DST-II) of this matrix.
float[][][]	elements() Returns the elements of this matrix.
void	fft3() Computes the 3D discrete Fourier transform (DFT) of this matrix.
DenseLargeFComplexMatrix3D	getFft2Slices() Returns new complex matrix which is the 2D discrete Fourier transform (DFT) of each slice of this matrix.
DenseLargeFComplexMatrix3D	getFft3() Returns new complex matrix which is the 3D discrete Fourier transform (DFT) of this matrix.
DenseLargeFComplexMatrix3D	getIfft2Slices(boolean scale) Returns new complex matrix which is the 2D inverse of the discrete Fourier transform (IDFT) of each slice of this matrix.
DenseLargeFComplexMatrix3D	getIfft3(boolean scale) Returns new complex matrix which is the 3D inverse of the discrete Fourier transform (IDFT) of this matrix.
float	getQuick(int slice, int row, int column) Returns the matrix cell value at coordinate [slice,row,column].
void	idct2Slices(boolean scale) Computes the 2D inverse of the discrete cosine transform (DCT-III) of each slice of this matrix.
void	idct3(boolean scale) Computes the 3D inverse of the discrete cosine transform (DCT-III) of this matrix.
void	idht2Slices(boolean scale) Computes the 2D inverse of the discrete Hartley transform (IDHT) of each slice of this matrix.
void	idht3(boolean scale) Computes the 3D inverse of the discrete Hartley transform (IDHT) of this matrix.
void	idst2Slices(boolean scale) Computes the 2D inverse of the discrete sine transform (DST-III) of each slice of this matrix.
void	idst3(boolean scale) Computes the 3D inverse of the discrete sine transform (DST-III) of this matrix.
void	ifft3(boolean scale) Computes the 3D inverse of the discrete Fourier transform (IDFT) of this matrix.
FloatMatrix3D	like(int slices, int rows, int columns) Construct and returns a new empty matrix of the same dynamic type as the receiver, having the specified number of slices, rows and columns.
void	setQuick(int slice, int row, int column, float value) Sets the matrix cell at coordinate [slice,row,column] to the specified value.

Methods inherited from class cern.colt.matrix.tfloat.impl.WrapperFloatMatrix3D

like2D, vectorize, viewColumn, viewColumnFlip, viewDice, viewPart, viewRow, viewRowFlip, viewSelection, viewSlice, viewSliceFlip, viewStrides

Methods inherited from class cern.colt.matrix.tfloat.FloatMatrix3D

aggregate, aggregate, aggregate, aggregate, assign, assign, assign, assign, assign, assign, assign, assign, assign, cardinality, copy, equals, equals, get, getMaxLocation, getMinLocation, getNegativeValues, getNonZeros, getPositiveValues, like, normalize, set, toArray, toString, viewSelection, viewSorted, zAssign27Neighbors, zSum

Methods inherited from class cern.colt.matrix.AbstractMatrix3D

checkShape, checkShape, columns, columnStride, index, rows, rowStride, size, slices, sliceStride, toStringShort

Methods inherited from class cern.colt.matrix.AbstractMatrix

ensureCapacity, isView, trimToSize

Methods inherited from class cern.colt.PersistentObject

clone

Methods inherited from class java.lang.Object

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

Constructor Detail

DenseLargeFloatMatrix3D

public DenseLargeFloatMatrix3D(int slices,
                       int rows,
                       int columns)

Method Detail

dct3

public void dct3(boolean scale)

Computes the 3D discrete cosine transform (DCT-II) of this matrix.

Overrides:

dct3 in class WrapperFloatMatrix3D

Parameters:

scale - if true then scaling is performed

dct2Slices

public void dct2Slices(boolean scale)

Computes the 2D discrete cosine transform (DCT-II) of each slice of this matrix.

Overrides:

dct2Slices in class WrapperFloatMatrix3D

Parameters:

scale - if true then scaling is performed

dht3

public void dht3()

Computes the 3D discrete Hartley transform (DHT) of this matrix.

Overrides:

dht3 in class WrapperFloatMatrix3D

dht2Slices

public void dht2Slices()

Computes the 2D discrete Hartley transform (DHT) of each slice of this matrix.

Overrides:

dht2Slices in class WrapperFloatMatrix3D

dst3

public void dst3(boolean scale)

Computes the 3D discrete sine transform (DST-II) of this matrix.

Overrides:

dst3 in class WrapperFloatMatrix3D

Parameters:

scale - if true then scaling is performed

dst2Slices

public void dst2Slices(boolean scale)

Computes the 2D discrete sine transform (DST-II) of each slice of this matrix.

Overrides:

dst2Slices in class WrapperFloatMatrix3D

Parameters:

scale - if true then scaling is performed

fft3

public void fft3()

Computes the 3D discrete Fourier transform (DFT) of this matrix. The physical layout of the output data is as follows:

 this[k1][k2][2*k3] = Re[k1][k2][k3]
                 = Re[(n1-k1)%n1][(n2-k2)%n2][n3-k3], 
 this[k1][k2][2*k3+1] = Im[k1][k2][k3]
                   = -Im[(n1-k1)%n1][(n2-k2)%n2][n3-k3], 
     0<=k1<n1, 0<=k2<n2, 0<k3<n3/2, 
 this[k1][k2][0] = Re[k1][k2][0]
              = Re[(n1-k1)%n1][n2-k2][0], 
 this[k1][k2][1] = Im[k1][k2][0]
              = -Im[(n1-k1)%n1][n2-k2][0], 
 this[k1][n2-k2][1] = Re[(n1-k1)%n1][k2][n3/2]
                 = Re[k1][n2-k2][n3/2], 
 this[k1][n2-k2][0] = -Im[(n1-k1)%n1][k2][n3/2]
                 = Im[k1][n2-k2][n3/2], 
     0<=k1<n1, 0<k2<n2/2, 
 this[k1][0][0] = Re[k1][0][0]
             = Re[n1-k1][0][0], 
 this[k1][0][1] = Im[k1][0][0]
             = -Im[n1-k1][0][0], 
 this[k1][n2/2][0] = Re[k1][n2/2][0]
                = Re[n1-k1][n2/2][0], 
 this[k1][n2/2][1] = Im[k1][n2/2][0]
                = -Im[n1-k1][n2/2][0], 
 this[n1-k1][0][1] = Re[k1][0][n3/2]
                = Re[n1-k1][0][n3/2], 
 this[n1-k1][0][0] = -Im[k1][0][n3/2]
                = Im[n1-k1][0][n3/2], 
 this[n1-k1][n2/2][1] = Re[k1][n2/2][n3/2]
                   = Re[n1-k1][n2/2][n3/2], 
 this[n1-k1][n2/2][0] = -Im[k1][n2/2][n3/2]
                   = Im[n1-k1][n2/2][n3/2], 
     0<k1<n1/2, 
 this[0][0][0] = Re[0][0][0], 
 this[0][0][1] = Re[0][0][n3/2], 
 this[0][n2/2][0] = Re[0][n2/2][0], 
 this[0][n2/2][1] = Re[0][n2/2][n3/2], 
 this[n1/2][0][0] = Re[n1/2][0][0], 
 this[n1/2][0][1] = Re[n1/2][0][n3/2], 
 this[n1/2][n2/2][0] = Re[n1/2][n2/2][0], 
 this[n1/2][n2/2][1] = Re[n1/2][n2/2][n3/2]
 

This method computes only half of the elements of the real transform. The other half satisfies the symmetry condition. If you want the full real forward transform, use getFft3. To get back the original data, use ifft3.

Overrides:

fft3 in class WrapperFloatMatrix3D

Throws:

IllegalArgumentException - if the slice size or the row size or the column size of this matrix is not a power of 2 number.

getFft2Slices

public DenseLargeFComplexMatrix3D getFft2Slices()

Returns new complex matrix which is the 2D discrete Fourier transform (DFT) of each slice of this matrix.

Overrides:

getFft2Slices in class WrapperFloatMatrix3D

Returns:

the 2D discrete Fourier transform (DFT) of each slice of this matrix.

getFft3

public DenseLargeFComplexMatrix3D getFft3()

Returns new complex matrix which is the 3D discrete Fourier transform (DFT) of this matrix.

Overrides:

getFft3 in class WrapperFloatMatrix3D

Returns:

the 3D discrete Fourier transform (DFT) of this matrix.

getIfft2Slices

public DenseLargeFComplexMatrix3D getIfft2Slices(boolean scale)

Returns new complex matrix which is the 2D inverse of the discrete Fourier transform (IDFT) of each slice of this matrix.

Overrides:

getIfft2Slices in class WrapperFloatMatrix3D

Parameters:

scale - if true then scaling is performed

Returns:

the 2D inverse of the discrete Fourier transform (IDFT) of each slice of this matrix.

getIfft3

public DenseLargeFComplexMatrix3D getIfft3(boolean scale)

Returns new complex matrix which is the 3D inverse of the discrete Fourier transform (IDFT) of this matrix.

Overrides:

getIfft3 in class WrapperFloatMatrix3D

Parameters:

scale - if true then scaling is performed

Returns:

the 3D inverse of the discrete Fourier transform (IDFT) of this matrix.

getQuick

public float getQuick(int slice,
             int row,
             int column)

Description copied from class: FloatMatrix3D

Returns the matrix cell value at coordinate [slice,row,column].

Provided with invalid parameters this method may return invalid objects without throwing any exception. You should only use this method when you are absolutely sure that the coordinate is within bounds. Precondition (unchecked): slice<0 || slice>=slices() || row<0 || row>=rows() || column<0 || column>=column().

Overrides:

getQuick in class WrapperFloatMatrix3D

Parameters:

slice - the index of the slice-coordinate.

row - the index of the row-coordinate.

column - the index of the column-coordinate.

Returns:

the value at the specified coordinate.

idct2Slices

public void idct2Slices(boolean scale)

Computes the 2D inverse of the discrete cosine transform (DCT-III) of each slice of this matrix.

Overrides:

idct2Slices in class WrapperFloatMatrix3D

Parameters:

scale - if true then scaling is performed

idht3

public void idht3(boolean scale)

Computes the 3D inverse of the discrete Hartley transform (IDHT) of this matrix.

Overrides:

idht3 in class WrapperFloatMatrix3D

Parameters:

scale - if true then scaling is performed

Throws:

IllegalArgumentException - if the slice size or the row size or the column size of this matrix is not a power of 2 number.

idht2Slices

public void idht2Slices(boolean scale)

Computes the 2D inverse of the discrete Hartley transform (IDHT) of each slice of this matrix.

Overrides:

idht2Slices in class WrapperFloatMatrix3D

Parameters:

scale - if true then scaling is performed

Throws:

IllegalArgumentException - if the slice size or the row size or the column size of this matrix is not a power of 2 number.

idct3

public void idct3(boolean scale)

Computes the 3D inverse of the discrete cosine transform (DCT-III) of this matrix.

Overrides:

idct3 in class WrapperFloatMatrix3D

Parameters:

scale - if true then scaling is performed

idst2Slices

public void idst2Slices(boolean scale)

Computes the 2D inverse of the discrete sine transform (DST-III) of each slice of this matrix.

Overrides:

idst2Slices in class WrapperFloatMatrix3D

Parameters:

scale - if true then scaling is performed

idst3

public void idst3(boolean scale)

Computes the 3D inverse of the discrete sine transform (DST-III) of this matrix.

Overrides:

idst3 in class WrapperFloatMatrix3D

Parameters:

scale - if true then scaling is performed

ifft3

public void ifft3(boolean scale)

Computes the 3D inverse of the discrete Fourier transform (IDFT) of this matrix. The physical layout of the input data has to be as follows:

 this[k1][k2][2*k3] = Re[k1][k2][k3]
                 = Re[(n1-k1)%n1][(n2-k2)%n2][n3-k3], 
 this[k1][k2][2*k3+1] = Im[k1][k2][k3]
                   = -Im[(n1-k1)%n1][(n2-k2)%n2][n3-k3], 
     0<=k1<n1, 0<=k2<n2, 0<k3<n3/2, 
 this[k1][k2][0] = Re[k1][k2][0]
              = Re[(n1-k1)%n1][n2-k2][0], 
 this[k1][k2][1] = Im[k1][k2][0]
              = -Im[(n1-k1)%n1][n2-k2][0], 
 this[k1][n2-k2][1] = Re[(n1-k1)%n1][k2][n3/2]
                 = Re[k1][n2-k2][n3/2], 
 this[k1][n2-k2][0] = -Im[(n1-k1)%n1][k2][n3/2]
                 = Im[k1][n2-k2][n3/2], 
     0<=k1<n1, 0<k2<n2/2, 
 this[k1][0][0] = Re[k1][0][0]
             = Re[n1-k1][0][0], 
 this[k1][0][1] = Im[k1][0][0]
             = -Im[n1-k1][0][0], 
 this[k1][n2/2][0] = Re[k1][n2/2][0]
                = Re[n1-k1][n2/2][0], 
 this[k1][n2/2][1] = Im[k1][n2/2][0]
                = -Im[n1-k1][n2/2][0], 
 this[n1-k1][0][1] = Re[k1][0][n3/2]
                = Re[n1-k1][0][n3/2], 
 this[n1-k1][0][0] = -Im[k1][0][n3/2]
                = Im[n1-k1][0][n3/2], 
 this[n1-k1][n2/2][1] = Re[k1][n2/2][n3/2]
                   = Re[n1-k1][n2/2][n3/2], 
 this[n1-k1][n2/2][0] = -Im[k1][n2/2][n3/2]
                   = Im[n1-k1][n2/2][n3/2], 
     0<k1<n1/2, 
 this[0][0][0] = Re[0][0][0], 
 this[0][0][1] = Re[0][0][n3/2], 
 this[0][n2/2][0] = Re[0][n2/2][0], 
 this[0][n2/2][1] = Re[0][n2/2][n3/2], 
 this[n1/2][0][0] = Re[n1/2][0][0], 
 this[n1/2][0][1] = Re[n1/2][0][n3/2], 
 this[n1/2][n2/2][0] = Re[n1/2][n2/2][0], 
 this[n1/2][n2/2][1] = Re[n1/2][n2/2][n3/2]
 

This method computes only half of the elements of the real transform. The other half satisfies the symmetry condition. If you want the full real inverse transform, use getIfft3.

Overrides:

ifft3 in class WrapperFloatMatrix3D

Parameters:

scale - if true then scaling is performed

Throws:

IllegalArgumentException - if the slice size or the row size or the column size of this matrix is not a power of 2 number.

setQuick

public void setQuick(int slice,
            int row,
            int column,
            float value)

Description copied from class: FloatMatrix3D

Sets the matrix cell at coordinate [slice,row,column] to the specified value.

Provided with invalid parameters this method may access illegal indexes without throwing any exception. You should only use this method when you are absolutely sure that the coordinate is within bounds. Precondition (unchecked): slice<0 || slice>=slices() || row<0 || row>=rows() || column<0 || column>=column().

Overrides:

setQuick in class WrapperFloatMatrix3D

Parameters:

slice - the index of the slice-coordinate.

row - the index of the row-coordinate.

column - the index of the column-coordinate.

value - the value to be filled into the specified cell.

elements

public float[][][] elements()

Description copied from class: FloatMatrix3D

Returns the elements of this matrix.

Overrides:

elements in class WrapperFloatMatrix3D

Returns:

the elements

like

public FloatMatrix3D like(int slices,
                 int rows,
                 int columns)

Description copied from class: FloatMatrix3D

Construct and returns a new empty matrix of the same dynamic type as the receiver, having the specified number of slices, rows and columns. For example, if the receiver is an instance of type DenseFloatMatrix3D the new matrix must also be of type DenseFloatMatrix3D, if the receiver is an instance of type SparseFloatMatrix3D the new matrix must also be of type SparseFloatMatrix3D, etc. In general, the new matrix should have internal parametrization as similar as possible.

Overrides:

like in class WrapperFloatMatrix3D

Parameters:

slices - the number of slices the matrix shall have.

rows - the number of rows the matrix shall have.

columns - the number of columns the matrix shall have.

Returns:

a new empty matrix of the same dynamic type.