[Home] [Back Issues] [Search] [Calendar] [Vol 1] [Vol 2] [Vol 3] [Vol 4] [Vol 5] [Vol 6]

The Wavelet Digest, Volume 6, Issue 3 (March 5, 1997)




Today's Editor: Wim Sweldens
                Lucent Technologies, Bell Laboratories
                wim@bell-labs.com



Today's Topics:

     1: Preprint: Understanding Wavelet Image Transform Codes
     2: Preprint: Preprints from H-Y Gao on shrinkage denoising
     3: Preprint: Wavelets in Numerical Analysis
     4: Preprint: Filter banks over commutative rings
     5: Preprint: Texture Matched Wavelet Filterbanks
     6: Preprint: Orthogonal Complex Filter Banks and Wavelets
     7: Preprint: Refinable distributions on Lie groups
     8: Preprint: Undecimated wavelet packet transform
     9: Preprint: A Wavelet Packet Model of Evoked Potentials
    10: Preprint: Wavelet Shrinkage of Poisson Data
    11: Thesis:   A scaling medium representation.
    12: Software: A Tcl/Tk tclet to generate wavelet functions
    13: Course:   Wavelets, Fuzzy Logic, Chaos, and Neural Networks
    14: Contents: Approximation Theory and its Applications 12-4
    15: Contents: Advances in Computational Mathematics 6-3,4 7-1,2
    16: Contents: J. Approx. Th. - feb97 
    17: Answer:   Wavelets and Communications (WD 6.2 #23 #24)
    18: Question: Periodic Noise Cancelling in Speech Signals
    19: Question: Atmospheric Boundary Layer Turbulence
    20: Question: Wavelets and edge detection
    21: Question: Smoothest Scaling Function
    22: Question: Denoising: How does N affect the threshold ?
    23: Question: Implementations of DPWT in C
     
Current number of members: 8947

E-mail:
   Help:                        help@wavelet.org
   Add yourself as a member:    add@wavelet.org
   Remove yourself as a member: remove@wavelet.org
   Publish in the next WD:      publish@wavelet.org

Visit us on the Web at http://www.wavelet.org/

Calendar of events:

 Jan 23-25: Workshop on Wavelets and Applications, Madras, India
 Mar 10-12: Workshop on Scientific Computing 97, Hong Kong
 Mar 20-22: Multiwavelets conference, Huntsville, TX
 Mar 21:    Wavelets & Subband transf. in Communications, NJIT
 Apr 13-16: 1997 ImageTech Conference Multimedia Imaging, Atlanta
 Apr 20   : Wavelet Transforms in Communications, Munich Germany
 Apr 22-24: SPIE Fourth Wavelet Application Conference, Orlando
*Jun 2-6:   Wavelets, Fuzzy Logic, Chaos, and Neural Networks, LA
 Jun 16-18: Engineering Applications of Neural Networks, Stockholm
 Jun 16-19: Sampling Theory and Applications, Aveiro, Portugal
 Jun 23-25: Multimedia Signal Processing Workshop, Princeton, NJ
 Jun 24-27: European conf. signal analysis and prediction, Prague
 Jun 25-27: Fractals in engineering, Arcachon, France
 Jul 2-4:   Scale-Space Theory in Computer Vision, Utrecht, Nl
 Jul 2-4:   Tutorial on Scale-Space Theory, The Netherlands
*Jul 21-27: Summer school: New interactions, Marseille, France
 Jul 25-Aug 1: SPIE Signal and Image Processing V, San Diego
 Aug 27-29: Time-Frequency and Time-Scale, Coventry, UK
 Sep 14-17: Time-Frequency Wavelet analysis symposium, Sacramento
 Sep 24-26: IEEE Neural Networks for Signal Processing, Florida
 Nov 16-Dec 6: UCLA short courses on communications, LA
 Dec 17-19: Introduction to wavelets, Boston



Date: Thu, 13 Feb 1997 11:08:24 +0100 (MET)
From: Stephane Mallat <mallat@cmapx.polytechnique.fr>
Subject: #1 Preprint: Understanding Wavelet Image Transform Codes

        Understanding Wavelet Image Transform Codes

        Stephane Mallat and Frederic Falzon 

        Ecole Polytechnique, CMAP, 91128 Palaiseau Cedex
        INRIA, Projet Pastis, Sophia Antipolis

                Abstract
The high resolution quantization hypothesis 
is not appropriate to evaluate the distortion-rate
of image transform codes in orthogonal bases.
It is shown that the compression performance depends mostly
on the precision of non-linear signal approximations with few non-zero bases 
vectors. At low bit rates, for a number of bit per pixel R
that is below 1, the distortion D(R) decays like $R^{-C}$ instead
of $2^{-2 R}$ as predicted by the high resolution transform
code theory. The constant C is typically of the order
of 1 for natural images decomposed in a wavelet basis,
or in a cosine basis for JPEG. The improved performance of embedded
codes in wavelet bases is also explained.

The paper is in the compressed file pub/Mallat/ImageCode.ps.gz that
can be retrieved by anonymous ftp at the address: 129.104.4.100

ftp://129.104.4.100/pub/Mallat/ImageCode.ps.gz



Date: Fri, 14 Feb 97 16:56 PST
From: gao@statsci.com (Hong-Ye Gao)
Subject: #2 Preprint: Preprints from H-Y Gao on shrinkage denoising

Preprints on wavelet shrinkage denoising are available by anonymous ftp:

ftp ftp.statsci.com
cd /pub/gao

Title: Wavelet Shrinkage Smoothing For Heteroscedastic Data
Author: Hong-Ye Gao
File: wsshd.ps.Z
Abstract:

   We extend Donoho and Johnstone's wavelet shrinkage smoothing
   technique (known as WaveShrink) to handle data with heteroscedastic
   noise.  We first show that if the noise variances are known,
   WaveShrink estimate achieves the same near-optimal convergence rate
   as in the white noise case.  We then propose a procedure for
   estimating the noise variances.  Our procedure is based on applying
   running MAD (Median Absolute Deviation from the median) to the
   non-decimated finest level wavelet coefficients.  We apply our
   technique to various numerical examples.
 
Title: Wavelet Shrinkage DeNoising Using Non-Negative Garrote
Author: Hong-Ye Gao
File: garrote.ps.Z
Abstract:

   In this paper, we combine Donoho and Johnstone's Wavelet Shrinkage
   denoising technique (known as WaveShrink) with Breiman's
   non-negative garrote.  We show that the non-negative garrote
   shrinkage estimate enjoys the same asymptotic convergence rate as
   the hard and the soft shrinkage estimates.  For finite sample
   simulations, non-negative garrote performs better (smaller
   mean-square-error) than both hard and soft, and comparable to the
   firm shrinkage.  We derive the minimax thresholds for the
   non-negative garrote.  We study the threshold selection procedure
   based on Stein's Unbiased Risk Estimate (SURE) for both
   non-negative garrote and soft shrinkages.

Title: Threshold Selection in WaveShrink
Author: Hong-Ye Gao
File: threshold.ps.Z
Abstract:

   Donoho and Johnstone's wavelet shrinkage denoising technique (known
   as WaveShrink) consists three steps: (1) transform data into
   wavelet domain; (2) shrink the wavelet coefficients; and (3)
   transform the shrunk coefficients back.  The choice of shrinkage
   function and thresholds in step (2) plays an important role for
   WaveShrink both theoretically and in practice.  In this paper, we
   discuss the issue of threshold selection in WaveShrink.  We first
   review the threshold selection procedure based minimax thresholds
   and Stein's Unbiased Risk Estimate (SURE).  We then propose a new
   threshold selection procedure based on combining Coifman and
   Donoho's cycle-spinning and SURE.  We call our new procedure
   SPINSURE.  We use examples to show that SPINSURE is numerically
   more stable than SURE: smaller standard deviation and smaller
   range.  Various comparisons with the ideal and minimax thresholds
   are also presented.



Date: Wed, 19 Feb 1997 12:15:57 +0100
From: Karsten Urban <urban@igpm.rwth-aachen.de>
Subject: #3 Preprint: Wavelets in Numerical Analysis

The following preprint is available under the site
ftp://www.igpm.rwth-aachen.de/pub/urban/reports/cdku.ps.Z

"Wavelets in Numerical Analysis and their Quantitative Properties"
                        by
        Wolfgang Dahmen, Angela Kunoth and Karsten Urban

Abstract: We discuss various instances where wavelets on the interval
serve as building blocks for extending wavelet methods to problems
that are neither periodic nor defined on the whole Euclidean space.
We briefly review then several properties of such wavelets which are
relevant for the previously mentioned applications. Finally, we take a
closer look at the quantitative stability properties of wavelets on
the interval and indicate ways of improving them.

                       Dr. Karsten Urban                          
  Institut fuer Geometrie und Praktische Mathematik, RWTH Aachen  
  Templergraben 55, 52056 Aachen, Germany                         
  e-Mail: urban@igpm.rwth-aachen.de                               
  Phone :  +49 / 241 / 80 63 38      Fax: +49 / 241 / 8888 317    



Date: Wed, 19 Feb 1997 16:25:29 +0100 (MET)
From: Andreas Klappenecker <klappi@ira.uka.de>
Subject: #4 Preprint: Filter banks over commutative rings

A new preprint is available:

Title:  Two-Channel Perfect Reconstruction Filter Banks over Commutative Rings
Authors: Andreas Klappenecker, Matthias Holschneider, and Kristin Flornes

Abstract: We develop a theory of perfect reconstructing filter banks
over commutative rings $A$ with identity. The filter bank viewpoint is
complemented by interpreting the signals as elements of the free
$A$-module of finitely supported sequences with values in $A.$ It is
shown that bases of this module can be obtained by the even translates
of the synthesis filter sequences of a perfect reconstructing filter
bank.  We associate a group structure with these bases and thereby
obtain a parametrization of synthesis filter pairs. It is proved that
this parametrization is complete, provided that $A$ is an arbitrary
field. As a special case we derive a complete parametrization of
biorthogonal real-valued filter pairs. We discuss lifting techniques
and their use to reduce the computational complexity of
implementations.

A copy can be downloaded from 
http://avalon.ira.uka.de/home/klappi/
or Email to
klappi@ira.uka.de

Thanks

Andreas Klappenecker
Universitaet Karlsruhe
IAKS
D-76128 Karlsruhe
Germany



Date: Wed, 19 Feb 1997 17:03:49 -0500
From: Thomas Greiner <TGREINER@compuserve.com>
Subject: #5 Preprint: Texture Matched Wavelet Filterbanks

Title: Orthogonal and Biorthogonal Texture Matched Wavelet Filterbanks for
Hierarchical Texture Analysis

Appeared in Signal Processing, No. 54, S. 1-22, Elsevier Verlag,
Niederlande, 1996

Author: Thomas Greiner

The paper deals with the application of the wavelet transform to
hierarchical texture analysis. The main goal is the design of
texture-matched FIR filterbanks as the dyadic (2 channel) orthogonal
and the M channel biorthogonal wavelet filterbanks. The approach
represents an extension of the mathematically-oriented design methods
for multirate and wavelet filterbanks. The cascaded application of the
filterbanks results in a critically, by the factor M × M (with M >=
2) downsampled, pyramidal representation of the textured image. The
performance of these new filters discriminating textures is compared
with the classical wavelet approaches on Brodatz´s textures.

A copy can be downloaded from
http://www.uni-kl.de/AG-Pandit/greiner.html
or Email to TGreiner@euromail.com 



Date: Tue, 25 Feb 1997 13:30:02 -0800
From: "Xiao-Ping ZHANG" <xiao@rome.eng.utsa.edu>
Subject: #6 Preprint: Orthogonal Complex Filter Banks and Wavelets

Dear Sir,

I would like to make the following preprint available to wavelet digest
readers:

Title: Orthogonal Complex Filter Banks and Wavelets: Some Properties
and Design

Authors: Xiao-Ping Zhang, Mita D. Desai and Ying-Ning Peng

Abstract: Recent wavelet research has primarily focused on real-valued
wavelet bases. However, complex wavelet bases offer a number of
potential advantages. For example, it has been recently suggested that
\cite{Lawton93, Lina94} that the complex Daubechies wavelet can be
made symmetric. However, these papers always imply that if the complex
basis has a symmetry property then it must exhibit linear phase as
well.  In this paper we prove that a linear phase complex orthogonal
wavelet does not exist. We study the implications of symmetry and
linear phase for both complex and real-valued orthogonal wavelet
bases. As a by-product, we propose a method to obtain a complex
orthogonal wavelet basis having the symmetry property and
approximately linear phase. The numerical analysis of the phase
response of various complex and real Daubechies wavelets is
given. Both real and complex-symmetric orthogonal wavelet can only
have symmetric amplitude spectra. Often it is desired to have
asymmetric amplitude spectra for processing general complex
signals. So we propose a method to design general complex orthogonal
perfect reconstruct filter banks (PRFBs) by parametrizing scheme.
Design examples are given. It is shown that the amplitude spectra of
the general complex conjugate quadraure filters (CQFs) can be
asymmetric with respect the zero frequency. This method can be used to
choose optimal complex orthogonal wavelet basis for processing complex
signals such as in radar and sonar.

Note: submitted to IEEE Trans. on Signal Processing, Special Issue on
Filter banks and wavelets.

A copy can be downloaded from
http://madrid.eng.utsa.edu/~xiao/papers/complex.ps
http://madrid.eng.utsa.edu/~xiao/papers/complex.ps.Z

or Email to xiao@madrid.eng.utsa.edu

Thanks

Sincerely,

Xiaoping Zhang



Date: Wed, 26 Feb 1997 16:25:32 +0800
From: wlawton@iss.nus.sg (Wayne Lawton)
Subject: #7 Preprint: Refinable distributions on Lie groups

Title: Infinite convolution products and refinable distributions on
       Lie groups

Wayne Lawton
Institute of System Science
National University of Singapore
Heng Mui Keng Terrace, Kent Ridge
Singapore 119597

Abstract: Sufficient conditions for the convergence in distribution of an
infinite convolution product $\mu_1*\mu_2*\ldots$ of measures on a
connected Lie group $\cG$ with respect to left invariant Haar measure
are derived. These conditions are used to construct distributions
$\phi$ that satisfy
        \mbox{$T\phi = \phi$}
where $T$ is a refinement operator constructed from a measure $\mu$
and a dilation automorphism $A$. The existence of $A$ implies $\cG$ is
nilpotent and simply connected and the exponential map is an analytic
homeomorphism. Furthermore, there exists a unique minimal compact
subset
        \mbox{$\cK \subset \cG$}
such that for any open set $\cU$ containing $\cK,$ and for any
distribution $f$ on $\cG$ with compact support, there exists an
integer $n(\cU,f)$ such that
        \mbox{$n \geq n(\cU,f)$}
implies
        \mbox{$\hbox{supp}(T^{n}f) \subset\cU.$}
If $\mu$ is supported on an $A$-invariant uniform subgroup $\gG,$ then
$T$ is related, by an intertwining operator, to a transition operator
$W$ on $\CC(\gG).$ Necessary and sufficient conditions for $T^{n}f$ to
converge to $\phi \in L^{2}$, and for the \mbox{$\gG$-translates} of
$\phi$ to be orthogonal or to form a Riesz basis, are characterized in
terms of the spectrum of the restriction of $W$ to functions supported
on
        \mbox{$\Omega := \cK \cK^{-1} \cap \gG.$}



Date: Mon, 3 Mar 97 13:31:52 GMT
From: a.walden@ic.ac.uk (A Walden)
Subject: #8 Preprint: Undecimated wavelet packet transform

The following preprint is available via WWW at:
http://www.ma.ic.ac.uk/statistics/links/atw.link/phase-corrected-TR-97-03.ps

Title: The phase-corrected undecimated discrete wavelet packet transform
and the recurrence of high latitude interplanetary shock waves.

[Imperial College of Science, Technology and Medicine, Statistics Section
Technical report TR-97-03, February 1997.]

Authors: Andrew T. Walden and Alberto Contreras Cristan

Abstract: We develop and apply advanced time-frequency methodology to
examine the recurrence time between shock waves identified in a
non-stationary time series of hourly-averaged southern hemisphere
solar magnetic field magnitude data acquired by the Ulysses
spacecraft. The discrete cyclic filtering steps of the maximal overlap
discrete wavelet packet transform (MODWPT) are fully explained. Energy
preservation is proven. With filter coefficients chosen from
Daubechies least asymmetric class, the optimum time shifts to apply to
ensure approximate zero phase filtering at every level of the MODWPT
are studied, and applied to the wavelet packet coefficients to give
phase-corrections which ensure alignment with the original time
series. Also the time series values at each time are decomposed into
details associated with each frequency band, and these line up
perfectly with features in the original time series since the details
are shown to arise through exact zero phase filtering. We carry out a
level 4 phase-corrected MODWPT of the Ulysses magnetic field data, and
show that the recurrence times of the shock waves previously
determined via manual pattern-matching on the raw data match those
times in the time-frequency plot where a broadband spectrum is
obtained; in other words, the phase-corrected MODWPT provides an
approach to picking the location of complicated events. Furthermore,
the phase-corrected MODWPT time-frequency plot strongly suggests that
the first shock wave is a composite of two events,
 possibly one associated with a  corotating interaction region, and one due 
to a coronal mass ejection. This might explain why the first shock wave has 
been differently classified in recent studies. 



Date: Tue, 4 Mar 1997 12:03:31 -0500 (EST)
From: Jonathan Raz <jonraz@sph.umich.edu>
Subject: #9 Preprint: A Wavelet Packet Model of Evoked Potentials

The following paper is available from:

http://www.sph.umich.edu/~jonraz/papers.html

or by contacting:

Jonathan Raz                        Telephone: (313) 936-1009
Department of Biostatistics         FAX: (313) 763-2215
University of Michigan              Email: jonraz@umich.edu
Ann Arbor, MI 48109-2029            

Title: A Wavelet Packet Model of Evoked Potentials

Authors: Jonathan Raz, Linda Dickerson, and Bruce Turetsky

Abstract: The standard methods for decomposition and analysis of
evoked potentials are bandpass filtering, identification of peak
amplitudes and latencies, and principal component analysis (PCA). We
discuss the limitations of these and other approaches, and introduce
wavelet packet analysis. Then we propose the ``single channel wavelet
packet model,'' a new approach in which a unique decomposition is
achieved using prior time/frequency information and differences in the
responses of the components to changes in experimental
conditions. Orthogonal sets of wavelet packets allow a parsimonious
time/frequency representation of the components. The method allows
energy in some wavelet packets to be shared among two or more
components, so the components are not necessarily orthogonal. The
single channel wavelet packet model and PCA both require constraints
to achieve a unique decomposition. In PCA, however, the constraints
are defined by mathematical convenience and may be unrealistic. In the
single channel wavelet packet model, the constraints are based on
prior scientific knowledge. We give an application of the method to
auditory evoked potentials recorded from cats. The good frequency
resolution of wavelet packets allows us to separate superimposed
components in these data. Our present approach yields estimates of
component waveforms and the effects of experiment conditions on the
amplitude of the components. We discuss future extensions that will
provide confidence intervals and p-values, allow for latency changes,
and represent multichannel data.



Date: Tue, 4 Mar 1997 14:02:48 -0600 (CST)
From: Eric Kolaczyk <kolaczyk@galton.uchicago.edu>
Subject: #10 Preprint: Wavelet Shrinkage of Poisson Data

The following three manuscripts, relating to de-noising Poisson data
with wavelets, are available via anonymous ftp.  The file format is 
compressed postscript (*.ps.Z)

1.)  Title: ``Estimation of Intensities of Burst-Like Poisson
              Processes Using Haar Wavelets.''
     Author:  Eric D. Kolaczyk

I present a method for producing estimates of the intensity function
of certain `burst-like' inhomogeneous Poisson processes, based on the
shrinkage of Haar wavelet coefficients of the observed counts.  The
Haar basis is a natural wavelet basis in which to work in this
context, and I derive thresholds for shrinkage estimation based on the
distribution of the coefficients.  The translation-invariant
de-noising approach of Donoho and Coifman (1995) is used in
conjunction with these level-dependent thresholds to yield smooth
estimates, without the usual `staircase' structure associated with
Haar wavelets.  This work is motivated by recent efforts in astronomy
to model the intensity functions underlying gamma-ray bursts.  It is
demonstrated that the method proposed herein (TIPSH) yields sharper
estimates of the detail structure in these signals than those obtained
through an analogous version of the standard adaptation of wavelet
shrinkage for Poisson counts, based on the square-root transformation.

ftp://galton.uchicago.edu/pub/kolaczyk/TIPSH_Mod.ps.Z

2.)  Title:  ``Non-Parametric Estimation of Gamma-Ray
               Burst Intensities Using Haar Wavelets.''
               The Astrophysical Journal, 1997  (to appear)
     Author:  Eric D. Kolaczyk

In this article, I present a method for the non-parametric
(model-free) estimation of intensity profiles underlying gamma-ray
bursts.  The method, TIPSH, is based on applying specially calibrated
thresholds to the Haar wavelet coefficients of binned counts gathered
from such bursts.  As functions well-localized with respect to both
time and scale, wavelets are an ideal tool for working with the often
sharp, abrupt nature of gamma-ray burst signals.  When applied to an
idealized signal in a small simulation study and a selection of actual
gamma-ray bursts, the TIPSH algorithm is found to be well capable of
simultaneously estimating the smooth, uniform background and the
pulse-like structure of gamma-ray burst signals.
 
ftp://galton.uchicago.edu/pub/kolaczyk/TIPSH_Appl.ps.Z

3.)  Title:  ``A Method for Wavelet Shrinkage Estimation of Certain
               Poisson Intensity Signals Using Corrected Thresholds.''
     Author:  Eric D. Kolaczyk

Wavelet shrinkage estimation has been found to be a powerful tool for
the non-parametric estimation of spatially variable phenomena.  Most
work in this area to date has concentrated primarily on the use of
wavelet shrinkage techniques in contexts where the data are modeled as
observations of a signal plus additive, Gaussian noise.  When the data
instead take the form of Poisson counts, a common procedure is to
first pre-process the data using Anscombe's square root
transformation, thereby normalizing the data and stabilizing the
variance.  However, this approach has a tendency to smooth away sharp,
brief structure in the underlying intensity function, especially in
situations involving very low levels of counts.  In this paper, I
introduce an alternative approach to estimating intensity functions
for a certain class of `burst-like' Poisson processes using wavelet
shrinkage.  The proposed method is based on the shrinkage of wavelet
coefficients of the original, un-transformed count data.  `Corrected'
versions of the usual Gaussian-based shrinkage thresholds are used.
The corrections explicitly account for effects of the first few
cumulants of the Poisson distribution on the tails of the coefficient
distributions.  A large deviations argument is used to justify these
corrections.  The performance of the new method is examined, and
compared to that of the pre-processing approach, in the context of an
application to an astronomical gamma-ray burst signal.

ftp://galton.uchicago.edu/pub/kolaczyk/PoisTholds.ps.Z

Communications regarding these manuscripts may be directed to
Eric Kolaczyk at kolaczyk@galton.uchicago.edu.



Date: Mon, 17 Feb 97 17:23:17 +0100
From: Felix Herrmann <felix@newdelphi.tn.tudelft.nl>
Subject: #11 Thesis: A scaling medium representation.

Hereby I would like to announce my thesis, entitled

   "A scaling medium representation, a discussion on well-logs,
    fractals and waves"

to all of Wavelet Digest readers with an interest in the use of
wavelets to tackle problems in physics. An electronic version of my
thesis can be obtained from my WWW page (without password):

   http://wwwak.tn.tudelft.nl/~felix/Thesispub/

where I included the gezipped POSTSCRIPT file PhDthesis.ps.gz.  There
are also hardcopy versions of my thesis available. However they are
limited in number. For those interested please give me, by email, your
address.

Kind regards,

Felix Herrmann.

SUMMARY

   The main subject of this research project was, and still is, to
   understand how the complexity of the earth's subsurface is
   transferred to the wavefield.  It was shown that the apparent
   complexity can be captured by means of a scaling medium
   representation where the H\"older exponents and the singularity
   spectrum constitute useful order of magnitude estimates. These
   estimates mathematically quantify the local and global aspects of
   the singularity structure delineating the scaling. At this point
   the question arises whether the current formulation for acoustic
   wave motion is capable of describing the transport of the
   singularity structure, displayed by the medium's heterogeneity, to
   the acoustic wavefield and vice versa.

KEYWORDS

   Scaling, Scale dynamics, Multifractals, Geophysics, Wave motion,
   Scattering, Wave localization, Homogenization theory, Seismic
   exploration, Multiscale analysis, Wavelets, Well-logs

Felix J. Herrmann
Lab. of Seismics and Acoustics    URL: http://wwwak.tn.tudelft.nl/~felix/
Centre for Technical Geoscience      e-mail: F.J.Herrmann@CTG.TUDelft.NL
Delft University of Technology          tel: +31 (0) 15 2783378
the Netherlands                         fax: +31 (0) 15 2783251



Date: Sat, 22 Feb 1997 13:59:30 -0600
From: "Mac A. Cody" <mcody@dfw.net>
Subject: #12 Software: A Tcl/Tk tclet to generate wavelet functions

Hi,

This is to announce the release of my new tclet, a wavelet function
generator.  A tclet (pronounced tickle-et) is a Tcl/Tk-based
application that can be run like a Java applet on a web browser.  It
requires the Sun Micosystems Tcl/Tk plugin for Netscape and Microsoft
browsers.  You can get the plugin at:
   http://www.sunlabs.com/research/tcl/plugin.

Here's what the tclet does:

1. Create orthonormal wavelets via the Pollen parameterization
algorithm, using two angular parameters, alpha and beta.

2. The values of the parameters can be selected by either a) typing
in the parameter values in two entry widgets, b) selecting a point
on a two-dimensional parameter space (on a canvas) with the mouse
pointer or the arrow keys, or c) pressing one of six buttons to
select predefined "interesting" parameter pairs.

3. Display the resulting wavelet coefficients generated by the
Pollen parameterization algorithm.

4. Display the recursively defined wavelet and scaling functions on
another canvas as generated by the "cascade" algorithm. Yes, it will
be faster when the Tcl/Tk plugin has the bytecode compiler in place!

All done in pure Tcl/Tk!  Go check it out at:

   http://www.dfw.net/~mcody/wavelet.html.

Don't forget to get and install a copy of the Tcl/Tk plugin first!

Also chek out my home page (listed below) for articles, software, and
pointers on the subject of wavelets.
 
Cheers,

Mac
-- 
Mac A. Cody                 Voice/Fax: 972-437-1517
408 Worcester Way           email: mcody@dfw.net
Richardson, TX 75080-3433   home page: http://www.dfw.net/~mcody



Date: Wed, 5 Mar 1997 08:39:00 -0800
From: "Goodin, Bill" <bgoodin@unex.ucla.edu>
Subject: #13 Course: Wavelets, Fuzzy Logic, Chaos, and Neural Networks

On June 2-6, 1997, UCLA Extension will present the short course,
"Wavelets, Fuzzy Logic, Chaos, and Neural Networks: Multiresolution
Soft-Computing Principles and Applications", on the UCLA campus in Los
Angeles.

The instructor is Harold Szu, PhD, Lamson Professor of Computer
Science, University of Southwestern Louisiana, Lafayette, and Past
President of the International Neural Network Society (INNS).

This course presents the multiresolution soft-computing principles and
applications of several different but related disciplines--wavelets,
neural networks, fuzzy logic, and chaos--in the context of
multiresolution granularity in human sensors system. This
"soft-computing" with words instead of yes-or-no binary logic is
useful for signal classification and compression, pattern recognition,
control of engineering tolerance imprecision, and the prediction of
fluctuating time series.

Various applications in chaos, fuzzy logic, wavelet transforms and
neural net learning are illustrated in terms of spatiotemporal
information processing, such as: signal/image de-noise; signal
classification; control device/machine chaos; communication coding;
chaotic heart and biomedical applications; intelligent bisensors such
as pacemakers and hearing aids; bomb 'sniffers'; and discrete wavelet
transform subband coding for image compression.

The course fee is $1495, which includes extensive course materials.
These materials are for participants only, and are not for sale.

For additional information and a complete course description, please
contact Marcus Hennessy at:

(310) 825-1047
(310) 206-2815  fax
mhenness@unex.ucla.edu
http://www.unex.ucla.edu/shortcourses

This course may also be presented on-site at company locations.



Date: Wed, 12 Feb 1997 15:09:24 +0100
From: Baltzer Science <mailer@ns.baltzer.nl>
Subject: #14 Contents: Approximation Theory and its Applications 12-4

Numerical Resolvent Methods for constrained Problems in Mechanics.
Gonzalo Alduncin. 1

Computation of Continuous Wavelet Transform ar Dyadic Scales by Subdivision
Scheme.
S. Riemenschneider and S. Xu. 26

Markov-Bernstein Type Inequalities of Multivariate Polynomials with Positive
Coefficients and Applications.
Lu Xuguang. 46

Wave Recursive Interpolation.
Ciu Zhenwen. 67

On Well-conditioned Boundary Value Problems for Systems of Second Order
Difference.
L. Jodar, E. Ponsoda and M. Legua Fernandez.  81

Approximation and Growth of Generalized Axisymmetric Potentials.
G.S. Srivastava. 96

A Quadrature Rule For Hadamard Finite Path Integrals.
Samir A. Ashour. 105

A Counterexample on Monotone Muentz Approximation.
S.P. Zhou. 111



Date: Fri, 14 Feb 1997 09:05:50 +0100
From: Baltzer Science <mailer@ns.baltzer.nl>
Subject: #15 Contents: Advances in Computational Mathematics 6-3,4 7-1,2

Advances in Computational Mathematics 6 (1996) 3-4

E. H. Twizell 
Preface 

Jack Howlett 
Foreword: JOHN CRANK 

J. C. Newby 
JOHN CRANK: his association with Brunel University 

J. Crank and P. Nicolson 
A practical method for numerical evaluation of solutions of partial differential
equations of the heat-conduction type 207-226 

P. C. Chatwin, D. M. Lewis and N. Mole 
Atmospheric diffusion: some new mathematical models 227-242 

G. Fairweather and J. C. Lopez-Marcos 
Galerkin methods for a semilinear parabolic problem with nonlocal boundary
conditions 243-262 

K. W. Morton and N. A. Burgess 
The stability of boundary conditions for an angled-derivative difference scheme
263-279 

J. R. Ockendon 
The role of the Crank-Gupta model in the theory of free and moving boundary
problems 281- 293 

K. A. Pericleous, M. Cross, G. Moran, P. Chow and K. S. Chan 
The role of the Crank-Gupta model in the theory of free and moving boundary
problems 295-308 

Simon Shaw and J. R. Whiteman 
Towards adaptive finite element schemes for partial differential Volterra equation
solvers 309-323 

J. H. Song, M. J. Edirisinghe and J. R. G. Evans 
Mathematical modelling of binder removal from plastically-formed engineering
ceramics 325-331 

E. H. Twizell, A. B. Gumel and M. A. Arigu 
Second-order, L_0-stable methods for the heat equation with time-dependent
boundary conditions 333-352 

D. A. Voss and A. Q. M. Khaliq 
Time-stepping algorithms for semidiscretized linear parabolic 
PDEs based on rational approximants with distinct real poles 353-363

Advances in Computational Mathematics 7 (1997) 1-2

Special Issue: Parallel methods for ODEs 

Editor: Kevin Burrage 

Kevin Burrage 
Parallel methods for ODEs 1-31 

Pierluigi Amodio and Luigi Brugnano 
Parallel ODE solvers based on block BVMs 5-26 

Claus Bendtsen 
A parallel stiff ODE solver based on MIRKs 27-36 

K. Burrage and H. Suhartanto 
Parallel iterated methods based on multistep Runge-Kutta methods of Radau type
37-57 

K. Burrage and H. Suhartanto 
Parallel iterated method based on multistep Runge-Kutta of Radau type for stiff
problems 59-77 

J. C. Butcher 
Order and stability of parallel methods for stiff problems 79-96 

Robert F. Enenkel and Kenneth R. Jackson 
DIMSEMs - diagonally implicit single-eigenvalue methods for the numerical solution
of stiff ODEs on parallel computers 97-133 

Min Hu, Ken Jackson, Jan Janssen and Stefan Vandewalle 
Remarks on the optimal convolution kernel for CSOR waveform relaxation 135-156 

P. J. van der Houwen and J. J. B. de Swart 
Parallel linear system solvers for Runge-Kutta methods 157-181 

P. J. van der Houwen and W. A. van der Veen 
Waveform relaxation methods for implicit differential equations 183-197 



Date: Fri, 28 Feb 1997 15:53:51 -0500 (EST)
From: Edit Kurali <kurali@math.ohio-state.edu>
Subject: #16 Contents: J. Approx. Th. - feb97 

Table of Contents: J. Approx. Theory, Volume 88, Number 2, February 1997

Song Li
Local smoothness of functions and Baskakov-Durrmeyer operators
139--153

Marek Be\'ska and Karol Dziedziul
Multiresolution approximation and Hardy spaces
154--167

Ferenc Weisz
Ces\'aro summability of two-parameter Walsh- Fourier series
168--192

Andi Kivinukk
On the measure of approximation for some linear means of trigonometric
   Fourier series
193--208

K. Zi\c{e}tak
Orthant-monotonic norms and overdetermined linear systems
209--227

Bernd Carl and Andreas Defant
Asymptotic estimates for approximation quantities of tensor product 
   identities
228--256

G. C. Kyriazis
Wavelet-type decompositions and approximations from shift-invariant 
   spaces
257--271

Note

Ding-Xuan Zhou
Extendibility of rational matrices
272--274



Date: Wed, 12 Feb 1997 13:24:14 -0500
From: "Alan R. Lindsey" <lindseya@rl.af.mil>
Subject: #17 Answer: Wavelets and Communications (WD 6.2 #23 #24)

Subject 23:
   Contact Richard Orr at Atlantic Aerospace.  His group (in conjunction
with Aware Corp.) have working prototype comm systems using
wavelet-based modulation.  I think CDMA is part of their wavelet-based
coding work.

Subject 24:
   See Lindsey and Medley, "Wavelets and Filter Banks in Digital
Communications," SPIE 1996, Orlando, Florida.  You can download this
paper, spie96.ps, via FTP as user "ftp" from voyager.cns.ohiou.edu in
directory pub/papers/lindsey.
   Richard Orr (see answer to question 23) also has an extensive
bibliography, which includes most, if not all, the above references as a
subset.  There are others as well but these should get you well on your way.

Alan R. Lindsey, Ph.D., P.E.        phone:  315-330-1879
Rome Laboratory/C3BB                  fax:  315-330-8013
525 Brooks Road                     email:  lindseya@rl.af.mil
Rome, NY  13441-4505



Date: Wed, 12 Feb 1997 10:17:56 -0800
From: "Hosking, Daniel S" <daniel.hosking@pss.boeing.com>
Subject: #18 Question: Periodic Noise Cancelling in Speech Signals

Hello, I'm searching for information related to sinusoidal noise
cancelling in speech signals.  I've been researching ways to control
feedback in PA systems without distorting the speech signal.  I've
tested adaptive LMS filters which haven't worked as well as expected -
they either converge too slowly or else they distort the speech
signal.  I'm hoping that there's been some work related to this
problem using wavelet transforms.  Any leads would be greatly
appreciated.

Thank you,
Daniel Hosking
email:  Daniel.Hosking@PSS.Boeing.com



Date: Tue, 18 Feb 1997 23:28:44 +0100
From: Christian Feigenwinter <feigenwinter@ubaclu.unibas.ch>
Subject: #19 Question: Atmospheric Boundary Layer Turbulence

Hello, I am a Ph.D. student working on the analysis of turbulence time
series in the atmospheric boundary layer measured with sonic
anemometers. I am looking for papers treating this subject with
respect to wavelets. I am also very interested in people working on
similar topics for an exchange of knowledge and experience. Thanks in
advance for your help, Christian Feigenwinter MCR Lab
Meteorology-Climatology-Remote Sensing Dep. of Geography Tel. +41 61
2726480 University of Basel Fax +41 61 2726923 Spalenring 145 CH-4055
Basel Christian Feigenwinter



Date: Tue, 18 Feb 1997 22:22:52 -0800
From: "Wendy Rinaldy<joerig@poboxes.com>" <wendyr@indosat.net.id>
Subject: #20 Question: Wavelets and edge detection

Hi

Recently I studied about wavelet. I want to ask if someone know about
aplication of WT on Edge Detection or Image Processing (specifically
on Segmentation). Please tell me if there are papers or books about
that.

Thanks for your attention
Sincerely yours

Wendy R (joerig@poboxes.com)



Date: Thu, 27 Feb 1997 10:56:15 -0600
From: Scott Gilbert <sgilbert@sunset.backbone.olemiss.edu>
Subject: #21 Question: Smoothest Scaling Function

What is the smoothest scaling function, w(x), solving a
'wavelet equation' of type:

(1) w(x) = c(0)w(2x) + c(1)w(2x-1) + c(2)w(2x-2) +...+ c(m)w(2x-m),

where the coefficients c(i) satisfy the 'usual conditions':

(2) sum c(i) over odd i = 1,

(3) sum c(i) over even i = 1,

and where w is normalized to have, say, integral = 1. 

The Schoenberg B-spline, with support [0,m], satisfies (1)-(3),
and I have found no smoother w's, after experimenting
with various c's, for m=3. Are B-splines the
smoothest (non-orthogonal) (father) wavelets?

[Answer]



Date: Fri, 28 Feb 1997 11:34:17 +1100 (EST)
From: andrewd@eng.uts.edu.au (Andrew Dorrell)
Subject: #22 Question: Denoising: How does N affect the threshold ?

Dear waveleteers,

I have been trying to apply the wavelet shrinkage techniques of Donoho
and Johnson to some practical image processing problems but keep
hitting a problem: the factor 'n' in the threshold calculation.  It
seems that the threshold at any level is dependent on the number of
coefficients (samples) you have!  This just doesn't make sense to me
(especially as I am considering the use of different sized "windows"
over a large image).  Surely the optimal threshold will be a function
of the noise variance only?  The distributions of the noise and signal
are assumed to be stationary aren't they?  Am I missing some
fundamental point in a big way?

Please help! :-/

Mr Andrew Dorrell School of Electrical Engineering University of
Technology, Sydney PO Box 123 Broadway NSW 2007 AUSTRALIA Phone: 61 2
330 2395 Fax: 61 2 330 2435 email: andrewd@ee.uts.edu.au
or dorrell@ihf.uts.edu.au



Date: Mon, 3 Mar 1997 15:42:19 -0500 (EST)
From: John Whitley <whitley@cs.buffalo.edu>
Subject: #23 Question: Implementations of DPWT in C

I am in need of C code for Neil Getz's Fast Discrete Periodic Wavelet
Transform (WD 2.1 topic#2).  It appears that Dr. Getz does not
presently have a C implementation of the DPWT.  I would like to know
if anyone has ported this algorithm to C, and is willing to share
their work.

Failing that, I shall undertake the port myself.  My port will be made
publicly available once completed.

-- John Whitley
   whitley@cs.buffalo.edu


<webmaster@wavelet.org>