Olga Koutchmy: Home Page

 
                   "A man can be destroyed, but not defeated" (Ernst Hemingway)
                   
                   
Koutchmy Olga
 Function : 
  
     Computer Engineer with the C.N.R.S.
 
 Address : 

     Laboratoire d'Analyse Numérique 
     Université Pierre et Marie Curie 
     Boite courrier 187
     75252 Paris Cedex 05 


 Phone  : 01.44.27.72.01 
 Fax    : 01.44.27.72.00 
 e-mail : koutchmy@ann.jussieu.fr

 Activities
 Publications

Activities

Participation in the project "Digital Image Processing," directed
  by Albert Cohen. Application of the wavelet thresholding method on
  images of special structures of astrophysical interest in the solar
  atmosphere, such as spicules.

In collaboration with Pascal Joly and within the framework of 
  the European TMR Project "Wavelets in Numerical Analysis and Simulation,"    
  administration of the Wavelet Database, dedicated to the software 
  written by various teams taking part in this project.

Within the framework of the  "Numerical Methods" workshop, directed
  by Christine Bernardi and Yvon Maday,  resolution of the
  Biharmonic Dirichlet problem using spectral type methods and
  collecting of a Package of programs relevant to spectral methods.

In collaboration with "Institut d'Astrophysique 
  de Paris (CNRS), of the McGill University, Montreal and the Institut
  IZMIRAN of Moscow researchers, modeling of magnetic fields to describe the 
  structures of Solar Corona.

Translation of several texts, from Russian to French, for example,
  "Evaluation of the Betti numbers" by Olga Oleinik, 
  "Bloc relaxation methods for two-dimensional elliptic equations" by Yuri Kouznetsov.

Other activities

Management of the Numerical Analysis Laboratory Library, in
 collaboration with Martine Legras, librarian, and Gérard Tronel, Professor.                                                                                            

Translation, from Russian to English,
  in collaboration with Gloria Girton, Professor of English, Paris,
  and Natalya V.Ivantchenko, Professor of Russian history at the
  University of Altaï, Russia, of the book "Wives of Ivan the Terrible:"
  About the book...

Selected publications

 
S.Koutchmy,E.Di Folco,F.Auchere,F.Boudin,J.-P.Delaboudiniere,
O.Koutchmy,J.-C.Noens,S.Rondi,R.Jimenez,R.N.Smartt
The Prolate Solar Chromosphere
SoHO8 Workshop (Paris), Kaldeich,B. ed., 1999, ESA SP
 
P.Lorrain and O.Koutchmy, 
The Sunspot as a Self-excited Dynamo.  
 Astronomy&Astrophysics, 339, 1998

P.Lorrain and O.Koutchmy, 
Horizontal magnetic forces on the photospheric granulation.  
( submitted to Solar Physics ), 1998 

S.L.Koutchmy, O.K.Koutchmy, V.A.Kutvitsky, M.M.Molodensky, and L.S.Soloviev,
Equilibrium and Stability of Axially Symmetrical Plasma Configurations in  
the Sun's Atmosphere.
Bulletin of the Russian Academy of Sciences, Physics, Vol. 59, No.9, pp.1570-1582,
1995. Copyright 1996 by Allerton Press,Inc

S.Koutchmy, V.Koutvitsky, M.Molodensky, L.Solov'iev and O.Koutchmy,
Magnetic Configurations of Coronal Streamers and Threads, 
Space Science Reviews, 70, 1994

O.Koutchmy, S.Koutchmy,
Optimum Filter and Frame Integration. Application to Granulation Pictures.
in 10th NSO/SPO Workshop Proceedings, O.von der Luhe,Ed. 1989

O.Koutchmy, S.Koutchmy, Ch.Nitschelm, J.Sykora and R.Smartt,
Image Processing of Coronal Pictures.
in Solar and Stellar Coronal Structure and Dynamics, R.Altrock,Ed. NSO/SPO 1988

Project "Digital Image Processing", directed by A.Cohen

  
 
    The topics developed are the following:
   
  - Image compression by adaptive quantification of multiscale 
    decompositions. 
    Analysis of the optimality of these algorithms in the framework of
    the information theory. 
    Use of wavelet coefficients tree structure to improve image encoding.

  - Image compression by decomposition, area by area (after segmentation 
    of the image).
    A patent application has been submitted to the CCETT 
    (Rennes). 
    Use of vector subbands filters for color images.

  - Noise removal by wavelet thresholding. 
    Analysis of the optimality of these algorithms in terms of "minimax" 
    of the non-parametric estimate.
    Study of the links existing between these methods and the techniques 
    of minimization of the functionals.
                                                                                     
  - This work has been carried out in collaboration with:

    D. Picard ( Université de Paris VII )
    F. Moreau (  CCETT, Rennes )
    J.P. d'Ales (  Ceremade, Université de Paris IX )
    R. DeVore ( University of South Carolina )
    I. Daubechies and M. Orchard ( University of Princeton )

Noise removal using wavelet thresholding technique. An application.

 The data is a temporal sequence of 15 gray level images showing the spicules
 above the solar chromosphere limb.
 Each image is 512 X 512 pixels.
 
 Note. The spicules are very fine jets of plasma launched in the corona,
 and these very fine details vary very quickly. 
 The origin of the spicules (small magnetic explosions?) and their roles in
 solar wind are unknown.
 
 The images underwent a degradation (aberrations) caused primarily by the
 inhomogeneities in the Earth's atmosphere: image motion,
 blur (loss of contrast), distortion.  There is no precise
 information on this degradation. We must restore these
 degraded images in order to improve contrast and the signal-to-noise
 ratio (noise removal), to highlight fine details (to locate the
 spatial high frequencies as well as possible), etc.  The 15 spicules images
 are of comparable quality.  But if we compute the
 average image, we see that the fine details are lost.
 For more details, see the notes in Tex:    
 "Examples of image processing."

 Spicules images Processing.

 Note to programs users.

 List of programs.

 References.

Model of 3D magnetic structures of solar corona.


 Technical report.

 Note to programs users.

 List of programs.

 References.

Last update: March 5, 2001.