[Home]
[Back Issues]
[Search]
[Calendar]
[Vol 1]
[Vol 2]
[Vol 3]
[Vol 4]
[Vol 5]
[Vol 6]
The Wavelet Digest, Volume 6, Issue 10 (October 1, 1997)
Today's Editor: Wim Sweldens
Lucent Technologies, Bell Laboratories
wim@bell-labs.com
Today's Topics:
1: Preprint: Periodic Cardinal Interpolatory Wavelets
2: Preprint: High-Dimensional Wavelet Data Compression
3: Preprint: Title: Integral Equations Using Wavelet Packets
4: Preprint: Preprints in wavelets and statistics from Guy Nason
5: Preprint: Wavelet Based Method for Nonlinear System Modeling
6: Preprint: Preprints on wavelet-based image compression
7: Preprint: Cross-correlation based on complex-wavelet transforms
8: Preprint: Balanced MultiWavelets - Theory and Design
9: Preprint: MRA and wavelet construction in R^n
10: Meeting: Proceedgins of the UK Time Freq Symposium available
11: Meeting: International Wavelet Conference in Morocco
12: Meeting: IEEE Workshop on multimedia compression
13: Job: PhD Research Position on Image Fusion
14: Job: Wavelets and dancing
15: Job: Post doctoral opening at Northeastern University
16: CFP: Special issue JMP on "Wavelets and Time-Frequency Analysis"
17: Contents: ETNA on CDROM
18: Contents: J. Approx. Th. - sep97 TOC
19: Contents: Numerical Algorithms 15-2
20: Answer: Methods to avoid shift variance of WT ? (WD 6.8 #25)
21: Question: Wavelets and Telecomm Engineering
22: Question: Wavelets and EMG signals from upper limbs.
23: Question: Biorthogonal wavelets and lifting scheme
24: Question: Segmentation based video coding.
25: Question: Choice of wavelet for image decomposition.
26: Question: Wavelets and convolution theorem
27: Question: daubwave.c and daubwave.doc from Steven Gollmer
Current number of members: 10159
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:
Oct 8-10: UCLA Course: Multirate Digital Filters, Los Angeles
Oct 12-13: Wavelets in statistics, Durham, NC
Nov 10-12: Course by Gil Strang in Braunschweig, Germany
Nov 16-Dec 6: UCLA short courses on communications, LA
Dec 8-12: Wavelets at Geophysics (GAU) Meeting, San Francisco
Dec 17-19: Introduction to wavelets, Boston
Jan 3-6: 9th Approximation Conference, Nashville, TN
*Apr 13-17: International Wavelet Conference, Tanger, Morocco
*Jun 22: IEEE Multimedia compression, Santa Barbara, CA
Date: Mon, 8 Sep 1997 09:50:24 -0500 (CDT)
From: Steven Xiao <slxiao@sulu.wustl.edu>
Subject: #1 Preprint: Periodic Cardinal Interpolatory Wavelets
Title: Periodic Cardinal Interpolatory Wavelets
Author:
Han-Lin Chen & Shaoliang Xiao
Abstract: In this paper, we construct periodic interpolating wavelets
and their duals from a periodic function $g(x)$ whose Fourier
coefficients are positive. The corresponding decomposition and
reconstruction algorithm is also given. The spline example shows that
such kind of wavelets shares good localization with any desired
regularity and symmetry. Our construction depends essentially on the
finite Fourier Transformation and the theory of circulant matrix.
From: Krasimir Kolarov <kolarov@interval.com>
Subject: #2 Preprint: High-Dimensional Wavelet Data Compression
Dear Wavelet Digest Readers,
There are a number of papers and reports on the topic of
High-Dimensional Wavelet Data Compression, from our
research group at Interval Research Corporation, available at
http://www.interval.com/~kolarov/pubs.html .
The list below includes papers published at DCC'97, SPIE'97,
Sequences'97 as well as Interval external technical reports. Some of
the abstracts are also included.
Communication and comments are greatly appreciated.
1.) Authors: Krasimir Kolarov and William Lynch
Title: "Compression of Functions Defined on Surfaces of 3D Objects"
Published: Data Compression Conference, J. Storer and M. Cohn, editors,
IEEE Computer Society Press, March 1997.
ABSTRACT
We present a technique to compress scalar functions defined on
2-manifolds. Our approach combines discrete wavelet transforms with
zerotree compression, building on ideas from three previous
developments: the lifting scheme, spherical wavelets, and embedded
zerotree coding methods. Applications lie in the efficient storage
and rapid transmission of complex data sets. Typical data sets are
earth topography, satellite images, and surface parametrizations. Our
contribution in this paper is the novel combination and application of
these techniques to general 2-manifolds.
2.) Authors: Krasimir Kolarov and Wiliiam Lynch
Title: "Optimization of the SW Algorithm for High-Dimensional Compression"
Published: Compression and Complexity of SEQUENCES 1997,
IEEE Computer Society Press, June 1997.
ABSTRACT
This paper describes an algorithm and a software package SW (Spherical
Wavelets) that implements a method for compression of scalar functions
defined on 3D objects. This method combines discrete second generation
wavelet transforms with an extension of the embedded zerotree coding
method. We present some results on optimizing the performance of the
SW algorithm via the use of arithmetic coding, different scaling and
norms of the wavelet coefficients. We describe an extension of the SW
algorithm using different prediction schemes in the zerotree
mechanism. The combined use of those techniques leads to a significant
improvement of the compression performance of SW.
3.) Authors: Krasimir Kolarov and William Lynch
Title: "Wavelet Compression for 3D and Higher-Dimensional Objects"
Published: Invited paper, Proc. of SPIE Conference on Applications of
Digital Image Processing, Volume 3164, San Diego,
California, July 1997.
4.)Authors: Thomas Yu, Krasimir Kolarov and William Lynch
Title: "Barysymmetric Multiwavelets on Triangle"
Published: Interval Research Corp. Technical Report IRC 1997-006,
March 11, 1997.
ABSTRACT:
Wavelets are typically designed on simple domains like
R<smaller>n</smaller> or rectangular subsets of
R<smaller>n</smaller>. Motivated by applications on more general
domains, for example, compression of data defined on the sphere or
more general surfaces, we tackle the problem of designing
multiresolution analysis on general manifolds.
In this paper, we give explicit construction of multiwavelets on
polygonal region in R<smaller>2</smaller> that is associated with a
nested triangular tessellation. Two different constructions will be
presented. The first construction is very similar to Alpert's
construction. Unlike the Alpert's 1-D construction, in which case
symmetry of basis functions comes in almost automatically, the
multiwavelets from our first construction possess no symmetry of any
sort. We define a form of symmetry for functions that "live" on
triangles, which we call barysymmetry, and establish various results
about it. We then show that by sacrificing some vanishing moments in
the first construction, we can construct multiwavelets which possess
barysymmetry - our second construction.
Typical members of the bases from both constructions have at least M>0
vanishing moments, but are discontinuous. We discuss how to apply
moment-interpolation schemes to improve these orthonormal bases, which
gives rise to their smooth biorthogonal counterparts.
5.) Authors: Krasimir Kolarov and William Lynch
Title: "Compression of Scalar Functions Defined on 2-Manifolds"
Published: Interval Research Corp. Technical Report 1996-012,
September 13, 1996.
Date: Fri, 12 Sep 1997 13:47:22 -0500 (CDT)
From: Hai Deng <hai@ling2.ece.utexas.edu>
Subject: #3 Preprint: Title: Integral Equations Using Wavelet Packets
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Title: Fast Solution of Electromagnetic Integral Equations Uing Adaptive
Wavelet Packet Transform
Authors: Hai Deng and Hao Ling
Department of Electrical and Computer Engineering
The University of Texas at Austin
Austin, TX 78712-1084
Abstract: The conventional wavelet transform (CWT) is widely used to
compress smooth operators. But it is not efficient for the compression
of some oscillatory operators, such as the Green function in
electromagnetic integral equations. In this paper we apply the
Adaptive Wavelet Packet Transform (AWPT) to the compression of moment
matrix, which is the discretized Green function, based on the
understanding that wavelet packet function is a modulated wavelet. A
cost function is employed to adaptively select the optimal wavelet
packet expansion/testing functions to achieve the maximum sparsity in
the transformed moment matrix. Numerical results show that the
thresheld elements in the AWPT transformed moment matrix grow only as
O(N^1.4) for typical scatterers. Consequently the solutions of the
transformed equations need only O(N^1.4) operations using recursive
methods such as Conjugate Gradient solver.
Electronic copies are available upon request (send email to
<hai@ling2.ece.utexas.edu>)
Date: Thu, 18 Sep 1997 15:46:46 +0100 (BST)
From: Guy Nason <g.p.nason@bristol.ac.uk>
Subject: #4 Preprint: Preprints in wavelets and statistics from Guy Nason
Preprint 1:
Title: Density and hazard rate estimation for right censored data using
wavelet methods.
Authors: Anestis Antoniadis and Girard Grigoire
Laboratoire LMC - IMAG,
Universiti Joseph Fourier
BP 53, 38041 Grenoble Cedex 09,
France
and
Guy Nason
Department of Mathematics
University of Bristol
University Walk, Bristol, BS8 1TW
England
Abstract: This paper describes a wavelet method for the estimation of density
and hazard rate functions from randomly right censored data. We adopt
a nonparametric approach in assuming that the density and hazard rate
have no specific parametric form. The method is based on dividing
the time axis into a dyadic number of intervals and then counting the
number of events within each interval. The number of events and the
survival function of the observations are then separately smoothed
over time via linear wavelet smoothers, and then the hazard rate
function estimators are obtained by taking the ratio. We prove that
the estimators possess pointwise and global mean square consistency,
obtain the best possible asymptotic MISE convergence rate and are
also asymptotically normally distributed. We also describe simulation
experiments that show these estimators are reasonably reliable
in practice. The method is illustrated with two real examples. The
first uses survival time data for patients with liver metastases from
a colorectal primary tumour without other distant metastases.
The second is concerned with times of unemployment for women and the
wavelet estimate, through its flexibility, provides a new and
interesting interpretation.
Keywords: survival data; hazard rate; wavelet estimation
To obtain the compressed PostScript:
http://www.stats.bris.ac.uk/pub/reports/Wavelets/whf02.ps.gz
Preprint 2.
Title: Statistical modelling of time series using non-decimated wavelet
representations
Authors: Guy Nason and Theofanis Sapatinas
Department of Mathematics, University of Bristol
University Walk, Bristol, BS8 1TW
England
and
Andrew Sawczenko
Institute of Child Health,
Royal Hospital for Sick Children,
Bristol, England.
Abstract: This article proposes the use of time-ordered non-decimated wavelet
or non-decimated wavelet packet coefficients to provide a
representation of a time series (explanatory). The resulting
representations are then used as variables in a statistical model
to provide predictions of another time series (response).
The statistical model provides valuable information about which
components in the explanatory time series drive the response time
series.
To represent our time series we use a collection of basis functions
known as wavelet packets. Our methodology modifies the standard
wavelet packets representation by providing an over-determined
representation of shifted wavelet packets where the shifts are not
restricted to the usual wavelet grid. We introduce a fast algorithm
for carrying out the time-ordered non-decimated wavelet packet
transform.
The shifted wavelet packet bases can represent many classes of time
series sparsely. The sparsity provides an effective dimension
reduction which enables standard statistical modelling techniques to
select the components of an explanatory series that best explain
the variation in the response time series.
Our modelling methodology is illustrated using time series examples
from two different arenas: (a) a biomedical example shows how infant
sleep states can be successfully classified using the time-ordered
non-decimated wavelet packet transform of heart rate and linear
discriminant analysis; (b) a renewable energy example using a
generalized linear model relating wind speeds at 40m to a
time-ordered non-decimated wavelet packet transform of wind speeds
at 10m.
Keywords: time series modelling; wavelets; dimension reduction; biomedical
time series; wind time series; linear discriminant analysis;
generalized linear models.
To obtain the compressed PostScript:
http://www.stats.bris.ac.uk/pub/reports/Wavelets/smotsunwr.ps.gz
See also
http://www.stats.bris.ac.uk/pub/reports/Wavelets
for other reports.
--
Guy Nason
School of Mathematics Phone: +44 (0117) 117 928 8633
University of Bristol FAX: +44 (0117) 117 928 7999
University Walk, Bristol, BS8 1TW, U.K.
Date: Friday, 19 September 1997
From: Wayne Lawton <wlawton@iss.nus.sg>
Subject: #5 Preprint: Wavelet Based Method for Nonlinear System Modeling
Authors:
Wayne Lawton
Department of Mathematics
National University of Singapore
10 Kent Ridge Crescent, Singapore 119260
Yi Yu
Centre for Information Enhanced Medicine
National University of Singapore
Kent Ridge, Singapore 119597
Seng Luan Lee
Department of Mathematics
National University of Singapore
10 Kent Ridge Crescent, Singapore 119260
Shaohua Tan
Department of Electrical Engineering
National University of Singapore
10 Kent Ridge Crescent, Singapore 119260
Abstract:
This paper formulates the nonlinear system modeling problem as a
scattered data interpolation problem, and develops a new method that
computes interpolants that minimize a wavelet-based norm subject to
interpolatory constraints. The norm is that of a Reproducing Kernel
Hilbert Space (RKHS) for which the wavelet functions that form an
orthonormal basis for L^2 are orthogonal. In contrast to radial basis
function kernels, these kernels are not translation invariant and they
may be designed to provide spatially varying resolution useful for
interpolating from unevenly distributed data samples. Furthermore, the
discrete wavelet transform can be exploited to efficiently compute the
values of the interpolant on a uniform grid. Numerical simulations are
used to demonstrate the method.
Date: Tue, 23 Sep 1997 12:25:26 +0200 (MESZ)
From: Detlev Marpe <d.marpe@fhtw-berlin.de>
Subject: #6 Preprint: Preprints on wavelet-based image compression
We would like to announce the availability of the following 3 papers on
wavelet-based image compression. You can download copies either from
our web-site
http://www.fhtw-berlin.de/Projekte/Wavelet
or via ftp from the URLs given below.
1) Title: Energy Constraint Scarce Wavelet Packet Libraries
for Image Compression
Authors: Detlev Marpe, Hans L. Cycon, and Wu Li
URL: ftp://ftp.rz.fhtw-berlin.de/pub/fhtw/fb3/pp4.ps.gz
Status: Preprint No. 541, Dept. of Mathematics,
Technical University Berlin, 1996.
Abstract: In this paper we introduce a generalization of the best basis
wavelet packet method which selects a suboptimal best basis out of a
restricted library of wavelet packet bases. For a given image this
restricted library is generated by a 2D separable multiresolution
analysis which discriminates branches of the resulting quadtree
structured representation of the image according to a measure of
energy-compaction. Thus we get a family of scarce libraries
(w.r.t. to the full wavelet packet library) parameterized by a
so-called energy-threshold. This energy-threshold parameter allows a
truncation of energetically irrelevant branches and hence a control
of the complexity of the best-basis algorithm ranging from the low
complexity standard DWT to the high complexity full best-basis
algorithm. Our experimental results for image coding applications
show that the rate-distortion (RD) performance increases in
distinguished (image dependent) jumps as the energy-threshold
decreases. As a consequence, the instrument of energy-thresholding
can be used to find optima with relatively high RD performance and
relatively low complexity.
2) Title: A Complexity Constraint Best-Basis Wavelet Packet Algorithm
for Image Compression
Authors: Detlev Marpe, Hans L. Cycon, and Wu Li
URL: ftp://ftp.rz.fhtw-berlin.de/pub/fhtw/fb3/ccbb6.ps.gz
Status: submitted to Applied and Computational Harmonic Analysis,
Academic Press, 1997.
Abstract: The concept of adapted waveform analysis using a best basis selection
out of a predefined library of wavelet packet (WP) bases allows an
efficient representation of a signal. These methods usually have the
disadvantage of high computational complexity. In this paper we
introduce an extension of the best-basis method, the complexity
constrained best-basis algorithm (CCBB) which allows an adaptive
approach, has relatively low complexity and is memory saving. Our
CCBB algorithm generates iteratively a scarce library of WP bases by
extending a given library according to the energy distribution of the
WP representation of the signal. This iteration process is terminated
by using a complexity measure as a control parameter. Our
experimental results for image coding applications show that the
rate-distortion (RD) performance increases in distinguished (image
dependent) jumps as the complexity is increased. This enables us to
find optima with relatively high RD performance and relatively low
complexity for processing still images as well as video sequences.
3) Title: Efficient Pre-Coding Techniques for Wavelet-Based Image Compression
Authors: Detlev Marpe and Hans L. Cycon
URL: ftp://ftp.rz.fhtw-berlin.de/pub/fhtw/fb3/pcs97.ps.gz
Status: Presented on Picture Coding Symposium, Berlin, Sept. 1997.
Abstract: The principle of transform coding is a successfully established
concept in image compression. In this paper we introduce a coding
method using a fast wavelet transform and an uniform quantizer
combined with a new framework of pre-coding techniques which are
based on the concepts of partitioning, aggregation and conditional
coding (PACC). Following these concepts, the data object emerging
from the quantizer will be first partitioned into different
subsources. Parts of correlations within and between different
subsources will then be captured by aggregating homogeneous elements
into data structures like run-length codes or zerotrees. By using
models based on conditional probabilities we are able to recover
correlations between the structures constructed before as well as
cross-correlations between different subsources which will be
utilized in a final arithmetic coding stage. Experimental results
show that our proposed coding methods have a rate-distortion (RD)
performance comparable to or even better than the best zerotree-based
still image coders in the published literature with the advantage of
a less demanding computational complexity. In addition, we propose
and evaluate a wavelet-based video coding algorithm which outperforms
the very efficient MPEG-4 Video Verification Model (VM 5.1) in both
subjective and objective quality.
Thanks.
Detlev Marpe
Wavelet Project
Fachhochschule fuer Technik u. Wirtschaft
Allee der Kosmonauten 20-22
10315 Berlin
GERMANY
e-mail: dmarpe@fhtw-berlin.de
Date: Wed, 24 Sep 1997 17:34:15 +0100
From: Guido Buresti <buresti@docenti.ing.unipi.it>
Subject: #7 Preprint: Cross-correlation based on complex-wavelet transforms
Dear Sirs,
I would like to inform you of two recent papers regarding a new
cross-correlation technique based on the complex-wavelet continuous
transform. The technique allows the time-local contribution of each
scale to the correlation of two signals to be characterized; the
time-local phase difference between the the fluctuations of the two
signals at each frequency may also be obtained. In the present papers
the technique is applied to the analysis of fluctuating velocity
signals derived in the first from the direct numerical solution of the
flow equations, and in the second from wind tunnel experiments.
For further information, please contact:
Prof. Guido Buresti
Dip. Ing. Aerospaziale - Universita' di Pisa
Via Diotisalvi, 2 - 56126 PISA - Italy
tel: +39-50-550200 fax: +39-50-553654
E-mail: buresti@ing.unipi.it =
Title:
Application of a wavelet cross-correlation analysis to DNS velocity
signals.
European Journal of Mechanics B/ Fluids, Vol. 16, N. 4, 1997, pp.
575-597
Authors:
M. Onorato Jr., M.V. Salvetti, G. Buresti, P. Petagna =
Abstract: A wavelet cross-correlation analysis is applied to the
velocity signals derived from axisymmetric direct numerical
simulations of a coaxial jet configuration. It is shown that this
technique permits to determine the time intervals and scales
contributing to the Reynolds stresses, and thus to assess the level of
intermittency of this contribution at the various scales. The
availability of the vorticity field from the numerical simulations
permits also to associate roll-up, passage and pairing of vortical
structures with average, spectral and instantaneous contributions to
the Reynolds stresses. High values of Reynolds stresses are found in
the roll-up region, where the axial and radial velocity fluctuations
have a small phase lag for significant time intervals. Conversely,
more downstream the passage of approximately symmetrical vortices,
although giving rise to high velocity fluctuations, does not produce
significant Reynolds stresses because the two velocity signals are
nearly orthogonal for most of the time. This demonstrates that
eduction schemes based on the energy of the velocity fluctuations
might not adequately reproduce the flow features characterizing the
Reynolds stress production.
Title:
Low aspect-ratio triangular prisms in cross-flow: measurements of the
wake fluctuating velocity field
Proc. 2nd European and African Conference on Wind Engineering, SGE Ed.,
1997, pp. 743-750.
Authors:
G. Buresti, G. Lombardi, A. Talamelli
Abstract: An experimental investigation is carried out to study the
characteristics of the fluctuating wake flow field of finite-length
prismatic bodies in cross-flow, placed vertically on a plane and
having isosceles triangular cross-section with 60=B0 or 90=B0 apex
angle orienta= ted in the downstream direction. The wake fluctuating
field is studied with hot-wire anemometry for aspect ratios h/w =3D
1.0, 1.5, 2.0 and 3.0 at Re==3D wU/n @ 1.2 x 105, and the velocity
signals are analysed by means of procedures based on the wavelet
transform. Velocity fluctuations with a clear dominating frequency are
found immediately outside the wake along all the models, even if their
magnitude decreases at a height corresponding to the free end of the
bodies. The wavelet analysis of the phase difference between signals
acquired at opposite sides of the wake shows that alternate vortex
shedding occurs for all the models. An increase in the phase shift
between the fluctuations on the same side of the wake with increasing
vertical distance between the probes suggests that the shed vortices
are curved. No evidence of symmetrical shedding or of the presence of
arch-type vortices is found, even for the models with the smaller
aspect ratios.
Date: Thu, 25 Sep 1997 13:34:57 +0200
From: "J. Lebrun" <lebrun@de.epfl.ch>
Subject: #8 Preprint: Balanced MultiWavelets - Theory and Design
The following preprint is available at the URL
http://lcavwww.epfl.ch/~lebrun/preprints/bmw_ieee98.ps.gz
Title: Balanced MultiWavelets - Theory and Design
Authors: Jerome Lebrun and Martin Vetterli
AudioVisual Communications Lab.
Communication Systems Division
Swiss Federal Institute of Technology
CH 1015 Lausanne, Switzerland
lebrun@de.epfl.ch or vetterli@de.epfl.ch
Abstract:
This paper deals with multiwavelets which are a recent
generalization of wavelets in the context of multirate filter banks
and with their applications to signal processing and especially
compression. By their inherent structure, multiwavelets are fit for
processing multi-channel signals. First, we will recall some general
results on multiwavelets and their implementations as multifilter
banks but also time-varying filter banks. Then, we will recall some
useful results on the convergence of the iterated matrix product
leading to them. Then, we will define under what conditions we can
apply systems based on multiwavelets to one-dimensional signals in a
simple way. That means we will give some natural and simple conditions
that should help in the design of new multiwavelets for signal
processing. Finally, we will provide some tools in order to construct
multiwavelets with the required properties, the so-called `balanced
multiwavelets'.
A longer version (technical report) is also available at
http://sscwww.epfl.ch/Pages/publications/ps_files/tr97_016.ps
Date: Tue, 30 Sep 1997 09:34:00 -0600 (MDT)
From: Lawrence Baggett <baggett@euclid.colorado.edu>
Subject: #9 Preprint: MRA and wavelet construction in R^n
Preprint:
Title:
Generalized Multiresolution Analyses, and
A Construction procedure for all Wavelet Sets in ${\Bbb R}^n.$
L. Baggett, H. Medina, and K. Merrill
Abstract: An abstract formulation of generalized multiresolution
analyses is presented, and those GMRA's that come from multiwavelets
are completely characterized. As an application of this abstract
formulation, a constructive procedure is developed, which produces all
wavelet sets in ${\Bbb R}^n$ relative to an integral, strict-dilation
matrix.
Date: Tue, 16 Sep 1997 15:48:15 +0000
From: "Stuart Lawson" <esrhy@eng.warwick.ac.uk>
Subject: #10 Meeting: Proceedgins of the UK Time Freq Symposium available
Proceedings of 2nd IEEE UK Symposium on Applications of Time-Frequency
and Time-Scale Methods held at University of Warwick, UK in August
1997 are available for purchase now at the cost of 20 UK pounds plus
postage and packing.
The symposium was sponsored by IEEE Signal Processing Chapter (UKRI
Section). The proceedings which runs to 192 pages plus 4 page insert,
has contributions in the application areas of Medicine, Image Analysis
and Filtering, Image Coding and Segmentation, Audio, Speech, etc.
Four keynote addresses are included by Gosta Granlund, Linkoping
University, Sweden; Pat Loughlin, Pittsburgh University, USA; David
Newland, Cambridge University, UK; Xavier Rodet, IRCAM, Paris, France.
For further information, please contact Dr Stuart Lawson at
Department of Engineering,
University of Warwick, Coventry, CV4 7AL, UK. Tel: +44 (0)
1203-523780, Fax: +44 (0)
1203-418922; e-mail: ssl@eng.warwick.ac.uk
Date: Fri, 26 Sep 1997 12:58:13 +0200 (MET DST)
From: Abdel Hak Ezzine <ezzine@cermics.enpc.fr>
Subject: #11 Meeting: International Wavelet Conference in Morocco
INTERNATIONAL WAVELET CONFERENCE
April 13-17, 1998
Tanger, Morocco
Call for papers
We are organizing an international conference on Wavelet and
Multiscale Methods at the Faculty of Sciences and Techniques of Tanger
in Morroco. The aim of this conference is the presentation of new
wavelet methods in numerical analysis, signal processing and
physics. In addition, this conference will be an opportunity for
maghrebin researchers to meet researchers working in the wavelet
subject.
The conference, organized as a one-week session, will be held at
Tanger, Morocco, and will take place 13-17 April 1998.
Preliminary list of invited Lectures
A. Arneodo ( Bordeaux-France ) "Ondelettes, fractales et phenomenes
multi-echelles : des concepts theoriques aux applications"
A. Cohen( Paris 6-France) "Adaptive multiscale methods for elliptic
problems"
W. Dahmen ( Aachen-Germany ) "Wavelets and Multigrid"
R. Coifman, (Participation to be confirmed)
I. Daubechies, (Participation to be confirmed)
R. DeVore, (Participation to be confirmed)
M. Farge ( ENS-France) Title to be announced
S. Jaffard ( Paris val de Marne) "Wavelets methods for multifractal
functions analysis"
J. Levy Vehel ( INRIA-France ) "Debruitage de signaux ou le calcul
de spectres multifractals"
Y. Meyer(CNRS-France ) Title to be announced
P. Tchamitchian ( Marseille-France ) "Analyse numerique des methodes
de Galerkin-ondelettes pour des problemes elliptiques"
V. Wickerhauser( Saint Louis-USA ) "Wavelets and Data Compression"
Conference Language:
The language for all activities of the conference will be English or French.
Submission of Contributions:
Abstracts of Papers and posters (1-2 pages) should be submitted until
November 30, 1997. Abstracts
will be subject to evaluation directly after submission and the
author will be informed about acceptance as soon as possible, but no
later than January 15, 1998.
Further information:
The second announcement for the meeting will be
mailed an September 30, 1997. It will contain information on
registration, accommodation, contributed talks and social events.
The Web site will be ready as soon as possible
http://cermics.enpc.fr/~ezzine/Tanger/conference.html
For more informations, please write to:
Ezzine Abdelhak Tel: 0164153532
ezzine@cermics.enpc.fr ENPC CERMICS
http://cermics.enpc.fr/~ezzine/home.html
Date: Mon, 29 Sep 1997 19:40:40 -0700 (PDT)
From: Roberto Manduchi <manduchi@robotics.stanford.edu>
Subject: #12 Meeting: IEEE Workshop on multimedia compression
CALL FOR PAPERS
IEEE WORKSHOP ON COMPUTER VISION TECHNIQUES FOR MULTIMEDIA COMPRESSION
Santa Barbara, CA - June 22, 1998
Sponsored by the PAMI Technical Committee of the IEEE Computer Society
The diffusion of multimedia has radically changed the traditional
notion of "video". Interactivity, panoramic rendering, hybrid 2D-3D,
characterize the presentation of visual contents on the WWW and
multimedia; videoconferencing is already the preferred medium for
distant communication; and large image and video databases are
becoming accessible and searchable.
The massive diffusion of interactive multimedia on the internet and on
CD-ROM calls for new, powerful compression architectures, that go well
beyond traditional waveform and block-based motion coding. Modern
compression standards such as MPEG-4 already support content-based
compression, making use of image understanding techniques,
traditionally developed in other contexts. Future techniques in
multimedia coding will consider issues such as the 3-D dynamic
modelling of objects, the semantic representation of the scene, the
interpolation of views for panoramic representations and mosaicing. It
is apparent that multimedia compression is becoming a challenging
application for computer vision.
The purpose of this workshop is to provide a forum for discussion of
new results on computer vision techniques for advanced scene analysis
and compression, as well as to encourage interaction with the image
compression community. The workshop will be held in conjunction with
CVPR'98. Workshop Proceedings will be available on site. Papers are
solicited for, but not limited to, the following areas:
- Layered representation of images
- Clustering of video shots
- Texture representation and segmentation for image coding
- 3-D segmentation and representation
- Compact combined description of shape and motion
- Registration of panoramic imagery
- Automatic facial conformation for model-based videophone
- Stereo and multiview image coding
- Parametric surface and volume description
- Speech-lip synchronization
- Contour-based coding
- Content-based scalability
PAPER SUBMISSION
Four copies of complete manuscript should be
received by Jan. 30, 1998 at the following address:
Dr. Carlo Tomasi - CVTMC'98
Department of Computer Science
Stanford University
Gates Building 1A, Room 156
Stanford, CA 94305
Papers should include:
(a) A title page containing the names and addresses of the
authors (including e-mail), an abstract of up to 200 words,
and one or more categories as listed above or other
keywords,
(b) A second title page - title and abstract only (to allow
for double blind reviewing),
(c) Paper - limited to 30 double-space pages (12 points,
1 inch margins), including figures, references, etc.
For more information, please check
http://www.vision.caltech.edu/manduchi/Workshop.html
Date: Fri, 19 Sep 1997 15:04:17 +0200
From: toet@sunglow.tm.tno.nl (Lex Toet)
Subject: #13 Job: PhD Research Position on Image Fusion
X-Sun-Charset: US-ASCII
PhD Research Position (4 years) on Image Fusion
TOPIC
The combination of images from different sensor modalities and with
different resolution into a single (more informative) image.
The Perception Department of the TNO Human Factors Research Institute
develops and evaluates algorithms to fuse images from different
electro- optical sensing modalities. Applications are target
detection and recognition, obstacle avoidance, surveillance and
navigation.
The aim of this project is to develop, implement and evaluate
algorithms to fuse images from multiple sensing modalities in a single
display with increased information content. The appearance of the
fused image should be invariant for time of day, and changes in
weather conditions and seasonal changes. The display should result in
an optimal task performance. The operational value of the algorithms
that are developed will be evaluated by measuring the task performance
of human obersvers performing detection, localisation, surveillance
and navigation tasks with fused imagery and in realistic scenario's.
The project is a collaboration between Signaal, TNO-HFRI and
University Utrecht.
WORK DESCRIPTION
The development and implementation of image fusion algorithms; the
design and performance of observer experiments to validate the newly
developed algorithms.
A training and educational program can be provided.
The work should result in a PhD thesis.
REQUIREMENTS
Candidates should be citizens of one of the NATO countries
and should have
- a degree in one of the exact sciences ,
- ample programming experience ,
- knowledge of and affinity with quantitative methods and techniques .
Interest in psychophysics is a prerequisite.
Experience in digital image processing is an advantage.
FURTHER INFORMATION
Dr. J.M. Valeton
Tel.: (+) 31-346-356239
Email: valeton@tm.tno.nl
Dr. A. Toet
Tel. : (+) 31-346-356237
Email: toet@tm.tno.nl
TNO Human Factors Research Institute
Kampweg 5
3769 DE Soesterberg
The Netherlands
Date: Wed, 17 Sep 1997 16:11:17 -0700
From: David Durlach <ddurlach@world.std.com>
Subject: #14 Job: Wavelets and dancing
I am a high-tech artist and inventor that build computer controlled
sculptures that dance with music. I and my design studio TechnoFrolics
are currently working on a project, funded in part by the National
Science Foundation, that includes work on developing DSP algorithms to
process music in order to permit our sculptures to dance in synchrony.
It is my hope to find a person (undergraduate, graduate student,
professor, or other) with an interest and expertise in Wavelet analysis
to collaborate with us on this project. A graduate student working on a
Masters or Ph.D. related to Wavelets might be ideal. We do not have the
financial resources to hire someone at normal industry wages, and
therefore frequently work with students in our area who get academic
credit for their work. Another possibility might be to find a professor
who wants his research shown off in a flashy manner.
For more information on me and my work, you may wish to examine a 5
page March 1997 article about TechnoFrolics on Discover Magazine's web
site http://www.enews.com/magazines/discover/magtxt/9703-5.html. It does
an excellent job at introducing our work.
Sincerely,
David Durlach
TECHNOFROLICS
11 Miller Street
Somerville, MA 02143
Phone 617-441-8870 Fax 617-441-8872
David Durlach Founder and Director
Date: Tue, 23 Sep 1997 13:28:22 -0400
From: "Eric L. Miller" <elmiller@ece.neu.edu>
Subject: #15 Job: Post doctoral opening at Northeastern University
Post-Doctoral Research Position Available at Northeastern University
in Computational Physics and Signal Processing.
Description: As part of a Multidisciplinary University Research
Initiative aimed at the detection, localization, and classification of
buried land mines, a post-doctoral research position is available at
Northeastern University in Boston, MA combining elements of
computational physics and signal/image processing. The research will
center around physics-based signal processing methods for solving the
demining problem. The initial phase of the work will be directed
toward the development of computationally efficient sensor models
describing the diffusion and/or propagation of energy through the
earth, the interaction of the energy with buried objects, and the
process of measuring the resulting scattered fields. In particular,
models for low frequency electromagnetic inductive systems and radio
frequency ground penetrating radar sensors are of interest. The
second portion of the research will center on the development of
signal and image processing algorithms based on these models for mine
detection, localization, and classification. Of interest here are
statistical signal processing methods employing elements of decision
and estimation theory which are robust to sensor noise and
environmental clutter. 20
Requirements: A Ph.D. and strong analytical skills in a field relevant
to the above described work (eg. Electrical Engineering, Mathematics,
Physics, Geoscience etc.) with experience in at least a subset of the
following areas:
1. Statistical signal processing,
2. Computational electromagnetics,
3. Numerical analysis
4. Inverse scattering,
5. Multiscale methods including wavelets
The candidate will be expected to carry out research in an independent
manner and if interested aid in the supervision of Master's and
Doctoral level graduate students. Strong oral and written English
skills are a must.
Funding is available for up to three years and I am looking for someone
to start as soon as possible.
All interested applicants are invited to contact
Prof. Eric Miller
235 Forsyth Building
Northeastern University
Boston, MA 02115
Tel: 617-373-838609
Email: elmiller@ece.neu.edu
Web http://www.cdsp.neu.edu/info/faculty/miller/miller.html
for more information or to submit an application (CV, references, and a
reprint of a published journal article).20
Date: Mon, 22 Sep 97 19:12:37 +0100
From: Bruno Torresani <torresan@cptsx1.univ-mrs.fr>
Subject: #16 CFP: Special issue JMP on "Wavelets and Time-Frequency Analysis"
ANNOUNCEMENT
JMP Special Issue
WAVELETS and TIME-FREQUENCY ANALYSIS
Dear waveleteers
I announced in the Wavelet Digest, Vol. 06, Nr. 01. (topic #14) the
publication of a special issue of the Journal of Mathematical Physics
on "Wavelets and time-frequency analysis". For technical reasons, the
publication date of this special issue has been postponed to the
summer 1998. Therefore, the deadline for submitting articles to this
issue is postponed to December 15, 1997.
As previously annouonced, half of the papers are invited papers, but
submissions are welcome.
For any information, please feel free to contact
Bruno Torr\'esani
CPT, CNRS-Luminy, Case 907
13288 Marseille Cedex 09
France
Phone: +33 4 91 26 95 23, Fax: +33 4 91 26 95 53
e-mail: torresan@cpt.univ-mrs.fr
Date: Sat, 13 Sep 1997 20:22:31 -0400 (EDT)
From: Lothar Reichel <reichel@mcs.kent.edu>
Subject: #17 Contents: ETNA on CDROM
ETNA on CDROM
The Electronic Transactions on Numerical Analysis (ETNA) was one of the
first completely electronic scientific journals when it was launched four
years ago.
ETNA has the following features:
* Manuscripts submitted to ETNA undergo peer review.
* Mathematical Reviews reviews all articles appearing in ETNA.
* Manuscripts may have color graphics.
* No annual page limitations makes fast publication possible.
* Key word search available.
* Forward referencing.
* ETNA can be accessed via the Internet by web browsers, ftp and e-mail.
* First four volumes of ETNA available on CDROM.
The CDROM version of ETNA can be accessed by a web browser (Mosaic).
Accessing ETNA in this manner is much more convenient than accessing ETNA
over the Internet when the Internet connection is slow. Many new PCs are
equipped with a CDROM reader. This makes the CDROM a convenient storage
medium for individuals and libraries.
The first ETNA CDROM contains papers with leading-edge contributions to
research in the following diverse areas of numerical analysis, analysis,
linear algebra and applications:
- conjugate gradient iterative methods and their variants,
for solving linear systems and for obtaining eigenvalues;
- general iterative methods for systems of linear equations,
including parallel synchronous and asynchronous multi-splitting and
inner-outer iteration methods;
- singular value and QR decompositions of matrices;
- applications of domain decomposition, multigrid, and finite
- element methods, and their preconditioning;
- Runge-Kutta methods, solutions of differential-algebraic
and Riccati equations in ordinary differential equation theory;
- numerical treatment of analysis problems, including Gaussian
quadrature, Pade' approximations and continued fractions, and the
Riemann Hypothesis.
- applications of numerical analysis, e.g., in signal processing and
control theory.
In addition the CDROM contains a variety of public domain software programs
(for Windows, Windows95, Sun Unix, SGI Unix, HP Unix, Digital Unix, and
Macintosh operating systems) that will allow you to display and print the
manuscripts included on the CDROM.
In order to defray the cost of producing the CDROM and to cover
shipping (by air mail), we are charging $25 for each CDROM. If you
wish to purchase the ETNA CDROM, please send $25.00 (money order,
travelers check, personal check for a U.S. bank or bank draft in U.S.
dollars) to
ETNA CDROM
Department of Mathematics and Computer Science
Kent State University
Kent, OH 44242
U. S. A.
For more information about ETNA and the ETNA CDROM send e-mail to
etna@mcs.kent.edu or regular mail to Professor A. Ruttan,
Department of Mathematics and Computer Science, Kent State University,
Kent, OH 44242, USA. ETNA's Internet address is http://etna.mcs.kent.edu.
Date: Mon, 22 Sep 1997 16:22:33 -0400 (EDT)
From: Edit Kurali <kurali@math.ohio-state.edu>
Subject: #18 Contents: J. Approx. Th. - sep97 TOC
Table of Contents: J. Approx. Theory, Volume 90, Number 3, September 1997
Robert Huotari and Wu Li
Continuities of metric projection and geometric consequences
319--339
Sergey K. Bagdasarov
Zolotarev $\omega$-polynomials in $W^rH^\omega[0,1]$
340--378
S. V. Konyagin
Divergence almost everywhere of a pointwise comparison of two sequences
of linear operators
379--384
Frank Deutsch, Wu Li, and Joseph D. Ward
A dual approach to constrained interpolation from a convex subset of
Hilbert space
385--414
Andr\'as Kro\'o and Darrell Schmidt
Some extremal problems for multivariate polynomials on convex bodies
415--434
Ammar Boukhemis
Study of sequence of classical orthogonal polynomials of dimension 2
435--454
Book reviews
455--466
Author index for Volume 90
467
Date: Tue, 23 Sep 1997 08:40:37 +0200
From: Baltzer Science <mailer@ns.baltzer.nl>
Subject: #19 Contents: Numerical Algorithms 15-2
Numerical Algorithms 15 (1997) 2
I.J. Anderson
A piecewise approach to piecewise approximation 139-152
Hans Strauss
On interpolation with products of positive definite functions 153-165
Celina Pestano-Gabino and Concepcion Gonzalez-Concepcion
Matrix Pade Approximation of rational functions 167-192
Vasile Dragan, Aristide Halanay and Adrian Stoica
Well-conditioned computation for H\infty controller near the optimum 193-206
N.M. Temme
Numerical algorithms for uniform Airy-type asymptotic expansions 207-225
Nicholas J. Higham
Stable iterations for the matrix square root 227-242
C. Brezinski
Book reviews 243-246
Date: Sun, 28 Sep 1997 10:19:54 +1000
From: Andrew Dorrell <andrewd@ee.uts.edu.au>
Subject: #20 Answer: Methods to avoid shift variance of WT ? (WD 6.8 #25)
Hi,
I have (perhaps correctly) been harassed for my breivety in answering
a previous question on shift invariance for wavelet transforms. Part
of my reason was that the majority of work I mensioned in my posting
could be found be searching the wavelet digest pages or the internet.
By way of explanation, I do not keep a detailed reference list on all
topics but feel that, if I know some theoretical issues and can pass
on the relevant keywords and search pointers then that may be of use
to the average researcher? However in acknowlegement of my previous
tersness, here is a follow up which will (hopefully) make my
(previous) answer more complete.
Geert Van der Auwera wrote:
>
> Dear Sir,
>
> Some weeks ago I sent you a mail containing the request for some
> references about techniques to overcome the shift-variance of the
> wavelet transformation. I think you are the ideal person to give
> me that information, because you answered a question about this
> subject in the previous Wavelet Digest. I have included it in this
> mail (see below). I would be very grateful if you could send me
> a reference on the following subjects :
>
> - the 'a-trous' algorithm
http://www.obspm.fr/service/sio/docs/html/doc/94NOV/vol2/node329.html
(thanks AltaVista)
Consider also:
[Rio.,Duh.'92] O.Rioul, P. Duhamel, "Fast Algorithms for Discrete and
Continuous Wavelet Transforms", IEEE Trans. on Inf. Theory, vol. 38,
no.2, March 1992,
[She.'92] M.J.Shensa, "The Discrete Wavelet Transform: Weding the A
Trous and Mallat Algorithms", IEEE Trans. on Signal
Proc. vol. 40, no.10, October 1992,
> - the web page of Eero Simoncelli
http://www.cis.upenn.edu/~eero/home.html
http://www.wavelet.org/wavelet/digest_04/digest_04.10.html#8
> - retaining odd and even samples
http://www.wavelet.org/wavelet/digest_06/digest_06.02.html#4
> - any other technique that might be important
Besides Daubechies book, the following talks about shift invariant
frames:
http://www.wavelet.org/wavelet/digest_06/digest_06.08.html#3
Some good references in the context of pattern recognition:
http://www.wavelet.org/wavelet/digest_05/digest_05.04.html#23
specifically:
R. R. Coifman and D. Donoho, Translation-invariant de-noising, in
Wavelets and Statistics (Eds. A. Antoniadis and G. Oppenheim),
Lecture Note in Statistics, Springer-Verlag, pp.125--150, 1995.
J. Liang and T. W. Parks, A translation-invariant wavelet representation
algorithm with applications, IEEE Trans. Sig. Proc., vol.44, no.2,
pp.225-- 232, 1996.
J.-C. Pesquet, H. Krim, and H. Carfantan, Time invariant orthonormal
wavelet representations, IEEE Trans. Sig. Proc., 1996, to appear.
Mr Andrew Dorrell
School of Electrical Engineering
University of Technology, Sydney
Date: Sat, 6 Sep 1997 16:45:51 +0000
From: Jorge Masdeu
Subject: #21 Question: Wavelets and Telecomm Engineering
Dear colleagues,
Since I first heard about wavelets last winter, I have been
trying hard to get familiar with the concept, then try to figure out
what the applications could be.To give an idea, I coded a "crude"
Daubechies 4 fwd and inv xform in my home computer that read "WAV"
files 8 bits PCM @ 11Khz sampling, and performed compression via
thresholding and produced reconstructed pseudo WAV output file to be
used for comparison. A real crude program that nevertherless measured
the SNR of the reconstructed inverse as well as the compression ratio.
Next item I want get acquinted with is wavelet packets, withpratical
"demos" on voice processing. The elusive point that I want to make
clear is that I am notparticulary interested in either Hilbert or
Besov spaces. Sometimes I wonder if there are readers as myself, more
geared towards pratical applications willing to set up an interest
group, sort of a "subset" of wavelet magazine readers,with a bias
toward practical telecomm applications. Those of you willing to share
experiences and "know-how" will be wellcome.
Date: Wed, 10 Sep 1997 08:34:34 +0200 (MET DST)
From: Pablo Saludes Rodil <pabsal@cantosan.tel.uva.es>
Subject: #22 Question: Wavelets and EMG signals from upper limbs.
Dear Wavelet Digest readers,
I am a undergraduate student at Valladolid University (Spain) and I am
developing my graduation project. I am interested in using wavelets for
analysing EMG signals from upper limbs. Could anybody suggest papers and
references that I can follow up? If anyone can help, please contact me.
Thanks in advance,
Pablo Saludes.
Pablo Saludes Rodil
e-mail: pabsal@cantosan.tel.uva.es
Undergraduate student at Valladolid University
Valladolid,
Spain
Date: Fri, 19 Sep 1997 16:41:23 +0200
From: Laurent Grisoni <grisoni@labri.u-bordeaux.fr>
Subject: #23 Question: Biorthogonal wavelets and lifting scheme
Dear Wavelet digest readers,
I have two questions about biorthogonal wavelets. The first one is the
mathematical interest about the vanishing moments of wavelets in the
biorthogonal case. The second one is how to make biorthogonal
symmetric wavelets with the lifting scheme, starting from lazy
wavelets.
Any informations would be greatly appreciated. Thanks in
advance for any indications.
Laurent Grisoni, PhD student
grisoni@labri.u-bordeaux.fr
LaBRI (Univ. Bordeaux 1),
351 cours de la liberation,
33405 TALENCE,
FRANCE
phone: (+33) 556-846-908
[Answer WD 6.11#20]
Date: Wed, 24 Sep 1997 10:45:52 -0700
From: "Mireya S. Garc€a V." <sgarcia@citedi.mx>
Subject: #24 Question: Segmentation based video coding.
Dear readers.
I want to know who can help me to find some about segmentation-based
video coding or software about that, or VLBR (mpeg4)
my e-mail is sgarcia@citedi.mx
Date: Mon, 29 Sep 1997 12:33:09 +0200
From: "Eipc" <assoing_eipc@netinfo.fr>
Subject: #25 Question: Choice of wavelet for image decomposition.
I have many questions which I hope some readers can provide me some
suggestion.
1) How can I choose a good or the better wavelet for an given image.
2) I have found many papers on texture analysis with wavelet, but nothing
about pattern recognition .What about that?
Thanks
D.Deguillemont
Tel : 33 3 21 38 85 10
Fax : 33 3 21 38 85 05
E-Mail : <ddeguillemont_eipc@netinfo.fr>
Date: Mon, 29 Sep 1997 12:41:39 -0400
From: Lang Withers <lwithers@psrw.com>
Subject: #26 Question: Wavelets and convolution theorem
Are there any wavelets that, in addition to having properties such as
(bi)orthogonality under translation and dilation, and the ability to
approximate piecewise linear functions, also have a fast convolution
algorithm (in either the "time"-domain or the transform domain)?
A familiar and important property of the (discrete) Fourier basis is
its (circular) convolution property, enabling fast convolution in the
transform domain by means of FFTs. So it's natural to ask if there
exist wavelets with a similar property exploitable by means of the
pyramid algorithm, or anyway with some kind of fast convolution.
Here is a possible hint:
(1) John Whelchel observed (personal communication) that for the
Hadamard matrix H and a symmetric Toeplitz matrix T, the matrix H'TH
has half its off-diagonal elements reduced to zero. Here I'm using H'
to denote the inverse of H.
(2) Compare this with the well known
result that for the DFT matrix F = (w^{jk}), (with w =exp(i2\pi/n) an
nth root of unity) and a cyclic matrix C, F'CF is diagonal. This
result follows from the relation FA=DF, where
D=diag(1,w,w^2,...,w^{n-1}) and A is the cyclic shift matrix. (Any
cyclic (circulant) matrix is a linear combination of successive powers
of A.)
(3) Similar to (1), we find for the Haar wavelet transform
matrix W, the matrix W'*C*W has all 0's in the first row and column
(except the 1,1 element) (Also true for W^T*C*W, with W^T the
transpose of W.) This is weaker than Whelchel's note (1), but closer
to the Fourier case (2) because it applies to cyclic
matrices. Moreover, for cyclic matrices made from arithmetic (evenly
spaced) sequences, such as C with first row [3,5,7,9], the entire
lower right quarter of W'*C*W is an identity matrix. (The Haar wavelet
W is the only one I've looked at.) -
Thank you.
Lang Withers
Pacific-Sierra Research Corporation
1400 Key Blvd., Suite 700, Arlington, VA 22209
phone:(703)807-5694
fax: (703)524-2420
email: lwithers@psrw.com
[Answer WD 6.11#21]
A similar question was posed in WD 4.2#38.
This generated two answers:[Answer WD 4.3#14] [Answer WD 4.4#13]
Date: Tue, 30 Sep 1997 13:14:11 +0200
From: Boyen Herman <hboyen@luc.ac.be>
Subject: #27 Question: daubwave.c and daubwave.doc from Steven Gollmer
Dear Wavelet Digest team,
When someone, like me, is interested in wavelets, then he or she can
find a lot of information in your Wavelet Digest. I do research to
receive a doctors degree in Applied Computer Science and my subject
was in november 1993 : "3D Medical Imaging". Since January of 1996, I'
m looking how I can use wavelets in the 3D and 4D Medical Imaging and
this led me to you. In the wavelet Digest - Volume 2, Issue 5(March
31, 1993) I've found an interesting program from Steven Gollmer :
daubwave.c with a good documentation : daubwave.doc. Now I've
translated it in Fortran 90 in a program daubwv1d.F90 wich do exact
the same as daubwave.c. I've tried to reach Steven Gollmer on
nls@mace.cc.purdue.edu, but the e-mail came back with the message
unknown Host. Is there anyone who know where I can find him now,
because I like to start whith his algorithm to expand the program, so
it can handle 4D-data and after that I will implemented this in my
Fortran90-code where I visualize the 2D and 3D medical images. Thank
you
Herman Boyen
Limburgs Universitair Centrum(L.U.C.)
Expertise centrum voor Digitale Media/ Labo Toegepaste Informatica
(EDM/LTI)
Wetenschapspark 2
B-3590 DIEPENBEEK
BELGIUM
Tel. +32 11 26.84.11
Fax. +32 11 26.84.00
PS This is not my full-time job. I'm assistent Computer Science at:
Hogeschool Limburg
Department IWT
Maastrichterstraat 100
B-3500 HASSELT
BELGIUM
Tel. +32 11 22.13.08
Fax. +32 11 23.18.07
<webmaster@wavelet.org>