@cindex DWT, see wavelet transforms @cindex wavelet transforms @cindex transforms, wavelet This chapter describes functions for performing Discrete Wavelet Transforms (DWTs). The library includes wavelets for real data in both one and two dimensions. The wavelet functions are declared in the header files @file{gsl_wavelet.h} and @file{gsl_wavelet2d.h}. @menu * DWT Definitions:: * DWT Initialization:: * DWT Transform Functions:: * DWT Examples:: * DWT References:: @end menu @node DWT Definitions @section Definitions @cindex DWT, mathematical definition The continuous wavelet transform and its inverse are defined by the relations, @iftex @tex \beforedisplay $$ w(s, \tau) = \int_{-\infty}^\infty f(t) * \psi^*_{s,\tau}(t) dt $$ \afterdisplay @end tex @end iftex @ifinfo @example w(s,\tau) = \int f(t) * \psi^*_@{s,\tau@}(t) dt @end example @end ifinfo @noindent and, @tex \beforedisplay $$ f(t) = \int_0^\infty ds \int_{-\infty}^\infty w(s, \tau) * \psi_{s,\tau}(t) d\tau $$ \afterdisplay @end tex @ifinfo @example f(t) = \int \int_@{-\infty@}^\infty w(s, \tau) * \psi_@{s,\tau@}(t) d\tau ds @end example @end ifinfo @noindent where the basis functions @c{$\psi_{s,\tau}$} @math{\psi_@{s,\tau@}} are obtained by scaling and translation from a single function, referred to as the @dfn{mother wavelet}. The discrete version of the wavelet transform acts on equally-spaced samples, with fixed scaling and translation steps (@math{s}, @math{\tau}). The frequency and time axes are sampled @dfn{dyadically} on scales of @math{2^j} through a level parameter @math{j}. @c The wavelet @math{\psi} @c can be expressed in terms of a scaling function @math{\varphi}, @c @c @tex @c \beforedisplay @c $$ @c \psi(2^{j-1},t) = \sum_{k=0}^{2^j-1} g_j(k) * \bar{\varphi}(2^j t-k) @c $$ @c \afterdisplay @c @end tex @c @ifinfo @c @example @c \psi(2^@{j-1@},t) = \sum_@{k=0@}^@{2^j-1@} g_j(k) * \bar@{\varphi@}(2^j t-k) @c @end example @c @end ifinfo @c @noindent @c and @c @c @tex @c \beforedisplay @c $$ @c \varphi(2^{j-1},t) = \sum_{k=0}^{2^j-1} h_j(k) * \bar{\varphi}(2^j t-k) @c $$ @c \afterdisplay @c @end tex @c @ifinfo @c @example @c \varphi(2^@{j-1@},t) = \sum_@{k=0@}^@{2^j-1@} h_j(k) * \bar@{\varphi@}(2^j t-k) @c @end example @c @end ifinfo @c @noindent @c The functions @math{\psi} and @math{\varphi} are related through the @c coefficients @c @c{$g_{n} = (-1)^n h_{L-1-n}$} @c @math{g_@{n@} = (-1)^n h_@{L-1-n@}} @c for @c{$n=0 \dots L-1$} @c @math{n=0 ... L-1}, @c where @math{L} is the total number of coefficients. The two sets of @c coefficients @math{h_j} and @math{g_i} define the scaling function and @c the wavelet. The resulting family of functions @c{$\{\psi_{j,n}\}$} @math{@{\psi_@{j,n@}@}} constitutes an orthonormal basis for square-integrable signals. The discrete wavelet transform is an @math{O(N)} algorithm, and is also referred to as the @dfn{fast wavelet transform}. @node DWT Initialization @section Initialization @cindex DWT initialization The @code{gsl_wavelet} structure contains the filter coefficients defining the wavelet and any associated offset parameters. @deftypefun {gsl_wavelet *} gsl_wavelet_alloc (const gsl_wavelet_type * @var{T}, size_t @var{k}) This function allocates and initializes a wavelet object of type @var{T}. The parameter @var{k} selects the specific member of the wavelet family. A null pointer is returned if insufficient memory is available or if a unsupported member is selected. @end deftypefun The following wavelet types are implemented: @deffn {Wavelet} gsl_wavelet_daubechies @deffnx {Wavelet} gsl_wavelet_daubechies_centered @cindex Daubechies wavelets @cindex maximal phase, Daubechies wavelets The is the Daubechies wavelet family of maximum phase with @math{k/2} vanishing moments. The implemented wavelets are @c{$k=4, 6, \dots, 20$} @math{k=4, 6, @dots{}, 20}, with @var{k} even. @end deffn @deffn {Wavelet} gsl_wavelet_haar @deffnx {Wavelet} gsl_wavelet_haar_centered @cindex Haar wavelets This is the Haar wavelet. The only valid choice of @math{k} for the Haar wavelet is @math{k=2}. @end deffn @deffn {Wavelet} gsl_wavelet_bspline @deffnx {Wavelet} gsl_wavelet_bspline_centered @cindex biorthogonal wavelets @cindex B-spline wavelets This is the biorthogonal B-spline wavelet family of order @math{(i,j)}. The implemented values of @math{k = 100*i + j} are 103, 105, 202, 204, 206, 208, 301, 303, 305 307, 309. @end deffn @noindent The centered forms of the wavelets align the coefficients of the various sub-bands on edges. Thus the resulting visualization of the coefficients of the wavelet transform in the phase plane is easier to understand. @deftypefun {const char *} gsl_wavelet_name (const gsl_wavelet * @var{w}) This function returns a pointer to the name of the wavelet family for @var{w}. @end deftypefun @c @deftypefun {void} gsl_wavelet_print (const gsl_wavelet * @var{w}) @c This function prints the filter coefficients (@code{**h1}, @code{**g1}, @code{**h2}, @code{**g2}) of the wavelet object @var{w}. @c @end deftypefun @deftypefun {void} gsl_wavelet_free (gsl_wavelet * @var{w}) This function frees the wavelet object @var{w}. @end deftypefun The @code{gsl_wavelet_workspace} structure contains scratch space of the same size as the input data and is used to hold intermediate results during the transform. @deftypefun {gsl_wavelet_workspace *} gsl_wavelet_workspace_alloc (size_t @var{n}) This function allocates a workspace for the discrete wavelet transform. To perform a one-dimensional transform on @var{n} elements, a workspace of size @var{n} must be provided. For two-dimensional transforms of @var{n}-by-@var{n} matrices it is sufficient to allocate a workspace of size @var{n}, since the transform operates on individual rows and columns. @end deftypefun @deftypefun {void} gsl_wavelet_workspace_free (gsl_wavelet_workspace * @var{work}) This function frees the allocated workspace @var{work}. @end deftypefun @node DWT Transform Functions @section Transform Functions This sections describes the actual functions performing the discrete wavelet transform. Note that the transforms use periodic boundary conditions. If the signal is not periodic in the sample length then spurious coefficients will appear at the beginning and end of each level of the transform. @menu * DWT in one dimension:: * DWT in two dimension:: @end menu @node DWT in one dimension @subsection Wavelet transforms in one dimension @cindex DWT, one dimensional @deftypefun int gsl_wavelet_transform (const gsl_wavelet * @var{w}, double * @var{data}, size_t @var{stride}, size_t @var{n}, gsl_wavelet_direction @var{dir}, gsl_wavelet_workspace * @var{work}) @deftypefunx int gsl_wavelet_transform_forward (const gsl_wavelet * @var{w}, double * @var{data}, size_t @var{stride}, size_t @var{n}, gsl_wavelet_workspace * @var{work}) @deftypefunx int gsl_wavelet_transform_inverse (const gsl_wavelet * @var{w}, double * @var{data}, size_t @var{stride}, size_t @var{n}, gsl_wavelet_workspace * @var{work}) These functions compute in-place forward and inverse discrete wavelet transforms of length @var{n} with stride @var{stride} on the array @var{data}. The length of the transform @var{n} is restricted to powers of two. For the @code{transform} version of the function the argument @var{dir} can be either @code{forward} (@math{+1}) or @code{backward} (@math{-1}). A workspace @var{work} of length @var{n} must be provided. For the forward transform, the elements of the original array are replaced by the discrete wavelet transform @c{$f_i \rightarrow w_{j,k}$} @math{f_i -> w_@{j,k@}} in a packed triangular storage layout, where @var{j} is the index of the level @c{$j = 0 \dots J-1$} @math{j = 0 ... J-1} and @var{k} is the index of the coefficient within each level, @c{$k = 0 \dots 2^j - 1$} @math{k = 0 ... (2^j)-1}. The total number of levels is @math{J = \log_2(n)}. The output data has the following form, @tex \beforedisplay $$ (s_{-1,0}, d_{0,0}, d_{1,0}, d_{1,1}, d_{2,0},\cdots, d_{j,k},\cdots, d_{J-1,2^{J-1} - 1}) $$ \afterdisplay @end tex @ifinfo @example (s_@{-1,0@}, d_@{0,0@}, d_@{1,0@}, d_@{1,1@}, d_@{2,0@}, ..., d_@{j,k@}, ..., d_@{J-1,2^@{J-1@}-1@}) @end example @end ifinfo @noindent where the first element is the smoothing coefficient @c{$s_{-1,0}$} @math{s_@{-1,0@}}, followed by the detail coefficients @c{$d_{j,k}$} @math{d_@{j,k@}} for each level @math{j}. The backward transform inverts these coefficients to obtain the original data. These functions return a status of @code{GSL_SUCCESS} upon successful completion. @code{GSL_EINVAL} is returned if @var{n} is not an integer power of 2 or if insufficient workspace is provided. @end deftypefun @node DWT in two dimension @subsection Wavelet transforms in two dimension @cindex DWT, two dimensional The library provides functions to perform two-dimensional discrete wavelet transforms on square matrices. The matrix dimensions must be an integer power of two. There are two possible orderings of the rows and columns in the two-dimensional wavelet transform, referred to as the ``standard'' and ``non-standard'' forms. The ``standard'' transform performs a complete discrete wavelet transform on the rows of the matrix, followed by a separate complete discrete wavelet transform on the columns of the resulting row-transformed matrix. This procedure uses the same ordering as a two-dimensional fourier transform. The ``non-standard'' transform is performed in interleaved passes on the rows and columns of the matrix for each level of the transform. The first level of the transform is applied to the matrix rows, and then to the matrix columns. This procedure is then repeated across the rows and columns of the data for the subsequent levels of the transform, until the full discrete wavelet transform is complete. The non-standard form of the discrete wavelet transform is typically used in image analysis. The functions described in this section are declared in the header file @file{gsl_wavelet2d.h}. @deftypefun {int} gsl_wavelet2d_transform (const gsl_wavelet * @var{w}, double * @var{data}, size_t @var{tda}, size_t @var{size1}, size_t @var{size2}, gsl_wavelet_direction @var{dir}, gsl_wavelet_workspace * @var{work}) @deftypefunx {int} gsl_wavelet2d_transform_forward (const gsl_wavelet * @var{w}, double * @var{data}, size_t @var{tda}, size_t @var{size1}, size_t @var{size2}, gsl_wavelet_workspace * @var{work}) @deftypefunx {int} gsl_wavelet2d_transform_inverse (const gsl_wavelet * @var{w}, double * @var{data}, size_t @var{tda}, size_t @var{size1}, size_t @var{size2}, gsl_wavelet_workspace * @var{work}) These functions compute two-dimensional in-place forward and inverse discrete wavelet transforms in standard and non-standard forms on the array @var{data} stored in row-major form with dimensions @var{size1} and @var{size2} and physical row length @var{tda}. The dimensions must be equal (square matrix) and are restricted to powers of two. For the @code{transform} version of the function the argument @var{dir} can be either @code{forward} (@math{+1}) or @code{backward} (@math{-1}). A workspace @var{work} of the appropriate size must be provided. On exit, the appropriate elements of the array @var{data} are replaced by their two-dimensional wavelet transform. The functions return a status of @code{GSL_SUCCESS} upon successful completion. @code{GSL_EINVAL} is returned if @var{size1} and @var{size2} are not equal and integer powers of 2, or if insufficient workspace is provided. @end deftypefun @deftypefun {int} gsl_wavelet2d_transform_matrix (const gsl_wavelet * @var{w}, gsl_matrix * @var{m}, gsl_wavelet_direction @var{dir}, gsl_wavelet_workspace * @var{work}) @deftypefunx {int} gsl_wavelet2d_transform_matrix_forward (const gsl_wavelet * @var{w}, gsl_matrix * @var{m}, gsl_wavelet_workspace * @var{work}) @deftypefunx {int} gsl_wavelet2d_transform_matrix_inverse (const gsl_wavelet * @var{w}, gsl_matrix * @var{m}, gsl_wavelet_workspace * @var{work}) These functions compute the two-dimensional in-place wavelet transform on a matrix @var{a}. @end deftypefun @deftypefun {int} gsl_wavelet2d_nstransform (const gsl_wavelet * @var{w}, double * @var{data}, size_t @var{tda}, size_t @var{size1}, size_t @var{size2}, gsl_wavelet_direction @var{dir}, gsl_wavelet_workspace * @var{work}) @deftypefunx {int} gsl_wavelet2d_nstransform_forward (const gsl_wavelet * @var{w}, double * @var{data}, size_t @var{tda}, size_t @var{size1}, size_t @var{size2}, gsl_wavelet_workspace * @var{work}) @deftypefunx {int} gsl_wavelet2d_nstransform_inverse (const gsl_wavelet * @var{w}, double * @var{data}, size_t @var{tda}, size_t @var{size1}, size_t @var{size2}, gsl_wavelet_workspace * @var{work}) These functions compute the two-dimensional wavelet transform in non-standard form. @end deftypefun @deftypefun {int} gsl_wavelet2d_nstransform_matrix (const gsl_wavelet * @var{w}, gsl_matrix * @var{m}, gsl_wavelet_direction @var{dir}, gsl_wavelet_workspace * @var{work}) @deftypefunx {int} gsl_wavelet2d_nstransform_matrix_forward (const gsl_wavelet * @var{w}, gsl_matrix * @var{m}, gsl_wavelet_workspace * @var{work}) @deftypefunx {int} gsl_wavelet2d_nstransform_matrix_inverse (const gsl_wavelet * @var{w}, gsl_matrix * @var{m}, gsl_wavelet_workspace * @var{work}) These functions compute the non-standard form of the two-dimensional in-place wavelet transform on a matrix @var{a}. @end deftypefun @node DWT Examples @section Examples The following program demonstrates the use of the one-dimensional wavelet transform functions. It computes an approximation to an input signal (of length 256) using the 20 largest components of the wavelet transform, while setting the others to zero. @example @verbatiminclude examples/dwt.c @end example @noindent The output can be used with the @sc{gnu} plotutils @code{graph} program, @example $ ./a.out ecg.dat > dwt.dat $ graph -T ps -x 0 256 32 -h 0.3 -a dwt.dat > dwt.ps @end example @iftex The graphs below show an original and compressed version of a sample ECG recording from the MIT-BIH Arrhythmia Database, part of the PhysioNet archive of public-domain of medical datasets. @sp 1 @center @image{dwt-orig,3.4in} @center @image{dwt-samp,3.4in} @quotation Original (upper) and wavelet-compressed (lower) ECG signals, using the 20 largest components of the Daubechies(4) discrete wavelet transform. @end quotation @end iftex @node DWT References @section References and Further Reading The mathematical background to wavelet transforms is covered in the original lectures by Daubechies, @itemize @asis @item Ingrid Daubechies. Ten Lectures on Wavelets. @cite{CBMS-NSF Regional Conference Series in Applied Mathematics} (1992), SIAM, ISBN 0898712742. @end itemize @noindent An easy to read introduction to the subject with an emphasis on the application of the wavelet transform in various branches of science is, @itemize @asis @item Paul S. Addison. @cite{The Illustrated Wavelet Transform Handbook}. Institute of Physics Publishing (2002), ISBN 0750306920. @end itemize @noindent For extensive coverage of signal analysis by wavelets, wavelet packets and local cosine bases see, @itemize @asis @item S. G. Mallat. @cite{A wavelet tour of signal processing} (Second edition). Academic Press (1999), ISBN 012466606X. @end itemize @noindent The concept of multiresolution analysis underlying the wavelet transform is described in, @itemize @asis @item S. G. Mallat. Multiresolution Approximations and Wavelet Orthonormal Bases of L@math{^2}(R). @cite{Transactions of the American Mathematical Society}, 315(1), 1989, 69--87. @end itemize @itemize @asis @item S. G. Mallat. A Theory for Multiresolution Signal Decomposition---The Wavelet Representation. @cite{IEEE Transactions on Pattern Analysis and Machine Intelligence}, 11, 1989, 674--693. @end itemize @noindent The coefficients for the individual wavelet families implemented by the library can be found in the following papers, @itemize @asis @item I. Daubechies. Orthonormal Bases of Compactly Supported Wavelets. @cite{Communications on Pure and Applied Mathematics}, 41 (1988) 909--996. @end itemize @itemize @asis @item A. Cohen, I. Daubechies, and J.-C. Feauveau. Biorthogonal Bases of Compactly Supported Wavelets. @cite{Communications on Pure and Applied Mathematics}, 45 (1992) 485--560. @end itemize @noindent The PhysioNet archive of physiological datasets can be found online at @uref{http://www.physionet.org/} and is described in the following paper, @itemize @asis @item Goldberger et al. PhysioBank, PhysioToolkit, and PhysioNet: Components of a New Research Resource for Complex Physiologic Signals. @cite{Circulation} 101(23):e215-e220 2000. @end itemize