jj2000.j2k.wavelet.analysis

Class AnWTFilterFloatLift9x7

java.lang.Object

jj2000.j2k.wavelet.analysis.AnWTFilter

jj2000.j2k.wavelet.analysis.AnWTFilterFloat

jj2000.j2k.wavelet.analysis.AnWTFilterFloatLift9x7

All Implemented Interfaces:

WaveletFilter

public class AnWTFilterFloatLift9x7
extends AnWTFilterFloat

This class inherits from the analysis wavelet filter definition for int data. It implements the forward wavelet transform specifically for the 9x7 filter. The implementation is based on the lifting scheme.

See the AnWTFilter class for details such as normalization, how to split odd-length signals, etc. In particular, this method assumes that the low-pass coefficient is computed first.

See Also:

AnWTFilter, AnWTFilterFloat

Field Summary

Fields

Modifier and Type	Field and Description
static float	ALPHA The value of the first lifting step coefficient
static float	BETA The value of the second lifting step coefficient
static float	DELTA The value of the fourth lifting step coefficient
static float	GAMMA The value of the third lifting step coefficient
private static float[]	HPSynthesisFilter The high-pass synthesis filter of the 9x7 wavelet transform
static float	KH The value of the high-pass subband normalization factor
static float	KL The value of the low-pass subband normalization factor
private static float[]	LPSynthesisFilter The low-pass synthesis filter of the 9x7 wavelet transform

Fields inherited from class jj2000.j2k.wavelet.analysis.AnWTFilter

OPT_PREFIX

Fields inherited from interface jj2000.j2k.wavelet.WaveletFilter

WT_FILTER_FLOAT_CONVOL, WT_FILTER_FLOAT_LIFT, WT_FILTER_INT_LIFT

Constructor Summary

Constructors

Constructor and Description
AnWTFilterFloatLift9x7()

Method Summary

Methods

Modifier and Type	Method and Description
void	analyze_hpf(float[] inSig, int inOff, int inLen, int inStep, float[] lowSig, int lowOff, int lowStep, float[] highSig, int highOff, int highStep) An implementation of the analyze_hpf() method that works on int data, for the forward 9x7 wavelet transform using the lifting scheme.
void	analyze_lpf(float[] inSig, int inOff, int inLen, int inStep, float[] lowSig, int lowOff, int lowStep, float[] highSig, int highOff, int highStep) An implementation of the analyze_lpf() method that works on int data, for the forward 9x7 wavelet transform using the lifting scheme.
boolean	equals(Object obj) Tests if the 'obj' object is the same filter as this one.
int	getAnHighNegSupport() Returns the negative support of the high-pass analysis filter.
int	getAnHighPosSupport() Returns the positive support of the high-pass analysis filter.
int	getAnLowNegSupport() Returns the negative support of the low-pass analysis filter.
int	getAnLowPosSupport() Returns the positive support of the low-pass analysis filter.
int	getFilterType() Returns the type of filter used according to the FilterTypes interface(W9x7).
float[]	getHPSynthesisFilter() Returns the time-reversed high-pass synthesis waveform of the filter, which is the high-pass filter.
int	getImplType() Returns the implementation type of this filter, as defined in this class, such as WT_FILTER_INT_LIFT, WT_FILTER_FLOAT_LIFT, WT_FILTER_FLOAT_CONVOL.
float[]	getLPSynthesisFilter() Returns the time-reversed low-pass synthesis waveform of the filter, which is the low-pass filter.
int	getSynHighNegSupport() Returns the negative support of the high-pass synthesis filter.
int	getSynHighPosSupport() Returns the positive support of the high-pass synthesis filter.
int	getSynLowNegSupport() Returns the negative support of the low-pass synthesis filter.
int	getSynLowPosSupport() Returns the positive support of the low-pass synthesis filter.
boolean	isReversible() Returns the reversibility of the filter.
boolean	isSameAsFullWT(int tailOvrlp, int headOvrlp, int inLen) Returns true if the wavelet filter computes or uses the same "inner" subband coefficient as the full frame wavelet transform, and false otherwise.
String	toString() Debugging method

Methods inherited from class jj2000.j2k.wavelet.analysis.AnWTFilterFloat

analyze_hpf, analyze_lpf, getDataType

Methods inherited from class jj2000.j2k.wavelet.analysis.AnWTFilter

getHPSynWaveForm, getLPSynWaveForm, getParameterInfo

Methods inherited from class java.lang.Object

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

Field Detail

LPSynthesisFilter

private static final float[] LPSynthesisFilter

The low-pass synthesis filter of the 9x7 wavelet transform

HPSynthesisFilter

private static final float[] HPSynthesisFilter

The high-pass synthesis filter of the 9x7 wavelet transform

ALPHA

public static final float ALPHA

The value of the first lifting step coefficient

See Also:

Constant Field Values

BETA

public static final float BETA

The value of the second lifting step coefficient

See Also:

Constant Field Values

GAMMA

public static final float GAMMA

The value of the third lifting step coefficient

See Also:

Constant Field Values

DELTA

public static final float DELTA

The value of the fourth lifting step coefficient

See Also:

Constant Field Values

KL

public static final float KL

The value of the low-pass subband normalization factor

See Also:

Constant Field Values

KH

public static final float KH

The value of the high-pass subband normalization factor

See Also:

Constant Field Values

Constructor Detail

AnWTFilterFloatLift9x7

public AnWTFilterFloatLift9x7()

Method Detail

analyze_lpf

public void analyze_lpf(float[] inSig,
               int inOff,
               int inLen,
               int inStep,
               float[] lowSig,
               int lowOff,
               int lowStep,
               float[] highSig,
               int highOff,
               int highStep)

An implementation of the analyze_lpf() method that works on int data, for the forward 9x7 wavelet transform using the lifting scheme. See the general description of the analyze_lpf() method in the AnWTFilter class for more details.

The coefficients of the first lifting step are [ALPHA 1 ALPHA].

The coefficients of the second lifting step are [BETA 1 BETA].

The coefficients of the third lifting step are [GAMMA 1 GAMMA].

The coefficients of the fourth lifting step are [DELTA 1 DELTA].

The low-pass and high-pass subbands are normalized by respectively a factor of KL and a factor of KH

Specified by:

analyze_lpf in class AnWTFilterFloat

Parameters:

inSig - This is the array that contains the input signal.

inOff - This is the index in inSig of the first sample to filter.

inLen - This is the number of samples in the input signal to filter.

inStep - This is the step, or interleave factor, of the input signal samples in the inSig array.

lowSig - This is the array where the low-pass output signal is placed.

lowOff - This is the index in lowSig of the element where to put the first low-pass output sample.

lowStep - This is the step, or interleave factor, of the low-pass output samples in the lowSig array.

highSig - This is the array where the high-pass output signal is placed.

highOff - This is the index in highSig of the element where to put the first high-pass output sample.

highStep - This is the step, or interleave factor, of the high-pass output samples in the highSig array.

See Also:

AnWTFilter.analyze_lpf(java.lang.Object, int, int, int, java.lang.Object, int, int, java.lang.Object, int, int)

analyze_hpf

public void analyze_hpf(float[] inSig,
               int inOff,
               int inLen,
               int inStep,
               float[] lowSig,
               int lowOff,
               int lowStep,
               float[] highSig,
               int highOff,
               int highStep)

An implementation of the analyze_hpf() method that works on int data, for the forward 9x7 wavelet transform using the lifting scheme. See the general description of the analyze_hpf() method in the AnWTFilter class for more details.

The coefficients of the first lifting step are [ALPHA 1 ALPHA].

The coefficients of the second lifting step are [BETA 1 BETA].

The coefficients of the third lifting step are [GAMMA 1 GAMMA].

The coefficients of the fourth lifting step are [DELTA 1 DELTA].

The low-pass and high-pass subbands are normalized by respectively a factor of KL and a factor of KH

Specified by:

analyze_hpf in class AnWTFilterFloat

Parameters:

inSig - This is the array that contains the input signal.

inOff - This is the index in inSig of the first sample to filter.

inLen - This is the number of samples in the input signal to filter.

inStep - This is the step, or interleave factor, of the input signal samples in the inSig array.

lowSig - This is the array where the low-pass output signal is placed.

lowOff - This is the index in lowSig of the element where to put the first low-pass output sample.

lowStep - This is the step, or interleave factor, of the low-pass output samples in the lowSig array.

highSig - This is the array where the high-pass output signal is placed.

highOff - This is the index in highSig of the element where to put the first high-pass output sample.

highStep - This is the step, or interleave factor, of the high-pass output samples in the highSig array.

See Also:

AnWTFilter.analyze_hpf(java.lang.Object, int, int, int, java.lang.Object, int, int, java.lang.Object, int, int)

getAnLowNegSupport

public int getAnLowNegSupport()

Returns the negative support of the low-pass analysis filter. That is the number of taps of the filter in the negative direction.

Returns:

2

getAnLowPosSupport

public int getAnLowPosSupport()

Returns the positive support of the low-pass analysis filter. That is the number of taps of the filter in the negative direction.

Returns:

The number of taps of the low-pass analysis filter in the positive direction

getAnHighNegSupport

public int getAnHighNegSupport()

Returns the negative support of the high-pass analysis filter. That is the number of taps of the filter in the negative direction.

Returns:

The number of taps of the high-pass analysis filter in the negative direction

getAnHighPosSupport

public int getAnHighPosSupport()

Returns the positive support of the high-pass analysis filter. That is the number of taps of the filter in the negative direction.

Returns:

The number of taps of the high-pass analysis filter in the positive direction

getSynLowNegSupport

public int getSynLowNegSupport()

Returns the negative support of the low-pass synthesis filter. That is the number of taps of the filter in the negative direction.

A MORE PRECISE DEFINITION IS NEEDED

Returns:

The number of taps of the low-pass synthesis filter in the negative direction

getSynLowPosSupport

public int getSynLowPosSupport()

Returns the positive support of the low-pass synthesis filter. That is the number of taps of the filter in the negative direction.

A MORE PRECISE DEFINITION IS NEEDED

Returns:

The number of taps of the low-pass synthesis filter in the positive direction

getSynHighNegSupport

public int getSynHighNegSupport()

Returns the negative support of the high-pass synthesis filter. That is the number of taps of the filter in the negative direction.

A MORE PRECISE DEFINITION IS NEEDED

Returns:

The number of taps of the high-pass synthesis filter in the negative direction

getSynHighPosSupport

public int getSynHighPosSupport()

Returns the positive support of the high-pass synthesis filter. That is the number of taps of the filter in the negative direction.

A MORE PRECISE DEFINITION IS NEEDED

Returns:

The number of taps of the high-pass synthesis filter in the positive direction

getLPSynthesisFilter

public float[] getLPSynthesisFilter()

Returns the time-reversed low-pass synthesis waveform of the filter, which is the low-pass filter. This is the time-reversed impulse response of the low-pass synthesis filter. It is used to calculate the L2-norm of the synthesis basis functions for a particular subband (also called energy weight).

The returned array may not be modified (i.e. a reference to the internal array may be returned by the implementation of this method).

Specified by:

getLPSynthesisFilter in class AnWTFilter

Returns:

The time-reversed low-pass synthesis waveform of the filter.

getHPSynthesisFilter

public float[] getHPSynthesisFilter()

Returns the time-reversed high-pass synthesis waveform of the filter, which is the high-pass filter. This is the time-reversed impulse response of the high-pass synthesis filter. It is used to calculate the L2-norm of the synthesis basis functions for a particular subband (also called energy weight).

The returned array may not be modified (i.e. a reference to the internal array may be returned by the implementation of this method).

Specified by:

getHPSynthesisFilter in class AnWTFilter

Returns:

The time-reversed high-pass synthesis waveform of the filter.

getImplType

public int getImplType()

Returns the implementation type of this filter, as defined in this class, such as WT_FILTER_INT_LIFT, WT_FILTER_FLOAT_LIFT, WT_FILTER_FLOAT_CONVOL.

Returns:

WT_FILTER_INT_LIFT.

isReversible

public boolean isReversible()

Returns the reversibility of the filter. A filter is considered reversible if it is suitable for lossless coding.

Returns:

true since the 9x7 is reversible, provided the appropriate rounding is performed.

isSameAsFullWT

public boolean isSameAsFullWT(int tailOvrlp,
                     int headOvrlp,
                     int inLen)

Returns true if the wavelet filter computes or uses the same "inner" subband coefficient as the full frame wavelet transform, and false otherwise. In particular, for block based transforms with reduced overlap, this method should return false. The term "inner" indicates that this applies only with respect to the coefficient that are not affected by image boundaries processings such as symmetric extension, since there is not reference method for this.

The result depends on the length of the allowed overlap when compared to the overlap required by the wavelet filter. It also depends on how overlap processing is implemented in the wavelet filter.

Parameters:

tailOvrlp - This is the number of samples in the input signal before the first sample to filter that can be used for overlap.

headOvrlp - This is the number of samples in the input signal after the last sample to filter that can be used for overlap.

inLen - This is the lenght of the input signal to filter.The required number of samples in the input signal after the last sample depends on the length of the input signal.

Returns:

true if both overlaps are greater than 2, and correct processing is applied in the analyze() method.

equals

public boolean equals(Object obj)

Tests if the 'obj' object is the same filter as this one. Two filters are the same if the same filter code should be output for both filters by the encodeFilterCode() method.

Currently the implementation of this method only tests if 'obj' is also of the class AnWTFilterFloatLift9x7

Overrides:

equals in class Object

Parameters:

obj - The object against which to test inequality.

getFilterType

public int getFilterType()

Returns the type of filter used according to the FilterTypes interface(W9x7).

Specified by:

getFilterType in class AnWTFilter

Returns:

The filter type.

See Also:

FilterTypes

toString

public String toString()

Debugging method

Overrides:

toString in class Object