anwtfilter.java

java 实现的小波压缩库代码,内部包含了分析器

/*
 * CVS identifier:
 *
 * $Id: AnWTFilter.java,v 1.15 2001/05/08 16:14:52 grosbois Exp $
 *
 * Class:                   AnWTFilter
 *
 * Description:             The abstract class for all analysis wavelet filters
 *
 *
 *
 * COPYRIGHT:
 * 
 * This software module was originally developed by Rapha雔 Grosbois and
 * Diego Santa Cruz (Swiss Federal Institute of Technology-EPFL); Joel
 * Askel鰂 (Ericsson Radio Systems AB); and Bertrand Berthelot, David
 * Bouchard, F閘ix Henry, Gerard Mozelle and Patrice Onno (Canon Research
 * Centre France S.A) in the course of development of the JPEG2000
 * standard as specified by ISO/IEC 15444 (JPEG 2000 Standard). This
 * software module is an implementation of a part of the JPEG 2000
 * Standard. Swiss Federal Institute of Technology-EPFL, Ericsson Radio
 * Systems AB and Canon Research Centre France S.A (collectively JJ2000
 * Partners) agree not to assert against ISO/IEC and users of the JPEG
 * 2000 Standard (Users) any of their rights under the copyright, not
 * including other intellectual property rights, for this software module
 * with respect to the usage by ISO/IEC and Users of this software module
 * or modifications thereof for use in hardware or software products
 * claiming conformance to the JPEG 2000 Standard. Those intending to use
 * this software module in hardware or software products are advised that
 * their use may infringe existing patents. The original developers of
 * this software module, JJ2000 Partners and ISO/IEC assume no liability
 * for use of this software module or modifications thereof. No license
 * or right to this software module is granted for non JPEG 2000 Standard
 * conforming products. JJ2000 Partners have full right to use this
 * software module for his/her own purpose, assign or donate this
 * software module to any third party and to inhibit third parties from
 * using this software module for non JPEG 2000 Standard conforming
 * products. This copyright notice must be included in all copies or
 * derivative works of this software module.
 * 
 * Copyright (c) 1999/2000 JJ2000 Partners.
 * */
package jj2000.j2k.wavelet.analysis;

import jj2000.j2k.codestream.writer.*;
import jj2000.j2k.wavelet.*;
import jj2000.j2k.image.*;
import jj2000.j2k.util.*;

import java.util.*;
import java.io.*;

/**
 * <p>This abstract class defines the methods of all analysis wavelet
 * filters. Specialized abstract classes that work on particular data types
 * (int, float) provide more specific method calls while retaining the
 * generality of this one. See the AnWTFilterInt and AnWTFilterFloat
 * classes. Implementations of analysis filters should inherit from one of
 * those classes.</p>
 *
 * All analysis wavelet filters should follow the following conventions:
 *
 * <ul>
 * <li>The first sample to filter is the low-pass one. As a consequence, if
 * the input signal is of odd-length then the low-pass output signal is one
 * sample longer than the high-pass output one. Therefore, if the length of
 * input signal is N, the low-pass output signal is of length N/2 if N is even
 * and N/2+1/2 if N is odd, while the high-pass output signal is of length N/2
 * if N is even and N/2-1/2 if N is odd.</li>
 *
 * <li>The normalization is 1 for the DC gain and 2 for the Nyquist gain (Type
 * I normalization), for both reversible and non-reversible filters.</li>
 *
 * <li>If the length of input signal is N, the low-pass output signal is of
 * length N/2 if N is even and N/2+1/2 if N is odd, while the high-pass output
 * sample is of length N/2 if N is even and N/2-1/2 if N is odd.</li>
 *
 * <li>The analyze method may seem very complicated, but is designed to
 * minimize the amount of data copying and redundant calculations when used
 * for block-based or line-based wavelet transform implementations, while
 * being applicable to full-frame transforms as well.</li>
 *
 * <li>All filters should implement the equals() method of the Object
 * class. The call x.equals(y) should test if the 'x' and 'y' filters are the
 * same or not, in what concerns the bit stream header syntax (two filters are
 * the same if the same filter code should be output to the bit stream).</li>
 * </ul>
 *
 * @see AnWTFilterInt
 * @see AnWTFilterFloat
 * */
public abstract class AnWTFilter implements WaveletFilter{

    /** The prefix for wavelet filter options: 'F' */
    public final static char OPT_PREFIX = 'F';

    /** The list of parameters that is accepted for wavelet filters. Options
     * for wavelet filters start with a 'F'. */
    private final static String [][] pinfo = {
        { "Ffilters", "[<tile-component idx>] <id> "+
          "[ [<tile-component idx>] <id> ...]",
          "Specifies which filters to use for specified tile-component. "+
          "If this option is not used, the encoder choses the filters "+
          " of the tile-components according to their quantization  type."+
          " If this option is used, a component transformation is applied "+
          "to the three first components.\n"+
          "<tile-component idx>: see general note\n"+
          "<id>: ',' separates horizontal and vertical filters, ':' separates"+
          " decomposition levels filters. JPEG 2000 part 1 only supports w5x3"+
          " and w9x7 filters.",null},
    };

    /**
     * Filters the input signal by this analysis filter, decomposing it in a
     * low-pass and a high-pass signal. This method performs the filtering and
     * the subsampling with the low pass first filtering convention.
     *
     * <p>The input signal resides in the inSig array. The index of the first
     * sample to filter (i.e. that will generate the first low-pass output
     * sample) is given by inOff. The number of samples to filter is given by
     * inLen. This array must be of the same type as the one for which the
     * particular implementation works with (which is returned by the
     * getDataType() method).</p>
     *
     * <p>The input signal can be interleaved with other signals in the same
     * inSig array, and this is determined by the inStep argument. This means
     * that the first sample of the input signal is inSig[inOff], the second
     * is inSig[inOff+inStep], the third is inSig[inOff+2*inStep], and so
     * on. Therefore if inStep is 1 there is no interleaving. This feature
     * allows to filter columns of a 2-D signal, when it is stored in a line
     * by line order in inSig, without having to copy the data, in this case
     * the inStep argument should be the line width.</p>
     *
     * <p>This method also allows to apply the analysis wavelet filter by
     * parts in the input signal using an overlap and thus producing the same
     * coefficients at the output. The tailOvrlp argument specifies how many
     * samples in the input signal, before the first one to be filtered, can
     * be used for overlap. Then, the filter instead of extending the input
     * signal will use those samples to calculate the first output
     * samples. The argument tailOvrlp can be 0 for no overlap, or some value
     * that provides partial or full overlap. There should be enough samples
     * in the input signal, before the first sample to be filtered, to support
     * the overlap. The headOvrlp provides the same functionality but at the
     * end of the input signal. The inStep argument also applies to samples
     * used for overlap. This overlap feature can be used for line-based
     * wavelet transforms (in which case it will only be used when filtering
     * the columns) or for overlapping block-based wavelet transforms (in
     * which case it will be used when filtering lines and columns).</p>
     *
     * <p>The low-pass output signal is placed in the lowSig array. The lowOff
     * and lowStep arguments are analogous to the inOff and inStep ones, but
     * they apply to the lowSig array. The lowSig array must be long enough to
     * hold the low-pass output signal.</p>
     *
     * <p>The high-pass output signal is placed in the highSig array. The
     * highOff and highStep arguments are analogous to the inOff and inStep
     * ones, but they apply to the highSig array. The highSig array must be
     * long enough to hold the high-pass output signal.</p>
     *
     * @param inSig This is the array that contains the input signal. It must
     * be of the correct type (e.g., it must be int[] if getDataType() returns
     * TYPE_INT).
     *
     * @param inOff This is the index in inSig of the first sample to filter.
     *
     * @param inLen This is the number of samples in the input signal to
     * filter.
     *
     * @param inStep This is the step, or interleave factor, of the input
     * signal samples in the inSig array. See above.
     *
     * @param tailOvrlp This is the number of samples in the input signal
     * before the first sample to filter that can be used for overlap. See
     * above.
     *
     * @param headOvrlp This is the number of samples in the input signal
     * after the last sample to filter that can be used for overlap. See
     * above.
     *
     * @param lowSig This is the array where the low-pass output signal is
     * placed. It must be of the same type as inSig and it should be long
     * enough to contain the output signal.
     *
     * @param lowOff This is the index in lowSig of the element where to put
     * the first low-pass output sample.
     *
     * @param lowStep This is the step, or interleave factor, of the low-pass
     * output samples in the lowSig array. See above.
     *
     * @param highSig This is the array where the high-pass output signal is
     * placed. It must be of the same type as inSig and it should be long
     * enough to contain the output signal.
     *
     * @param highOff This is the index in highSig of the element where to put
     * the first high-pass output sample.
     *
     * @param highStep This is the step, or interleave factor, of the
     * high-pass output samples in the highSig array. See above.
     *
     * @see WaveletFilter#getDataType
     * */
    public abstract 
        void analyze_lpf(Object inSig, int inOff, int inLen, int inStep,
                     Object lowSig, int lowOff, int lowStep,
                     Object highSig, int highOff, int highStep);
                     
    /**
     * Filters the input signal by this analysis filter, decomposing it in a
     * low-pass and a high-pass signal. This method performs the filtering and
     * the subsampling with the high pass first filtering convention.
     *
     * <p>The input signal resides in the inSig array. The index of the first
     * sample to filter (i.e. that will generate the first high-pass output
     * sample) is given by inOff. The number of samples to filter is given by
     * inLen. This array must be of the same type as the one for which the
     * particular implementation works with (which is returned by the
     * getDataType() method).</p>
     *
     * <p>The input signal can be interleaved with other signals in the same
     * inSig array, and this is determined by the inStep argument. This means
     * that the first sample of the input signal is inSig[inOff], the second
     * is inSig[inOff+inStep], the third is inSig[inOff+2*inStep], and so
     * on. Therefore if inStep is 1 there is no interleaving. This feature