AG Numerik

The local HASSIP page at Marburg

HASSIP is the acronym for "Harmonic Analysis and Statistics for Signal and Image Processing purposes", the title of a European Union RTN network .

Presentation

HASSIP is a Research Training Network funded by the European Commission within the programme Improving the Human Potential. As such, it will develop research and training activities in various areas of mathematical analysis and statistics, related to signal and image processing problems. HASSIP will also offer job opportunities (Marie Curie doctoral and postdoctoral fellowships) in its sites.

Scientific Objectives

The aim of the HASSIP network is to develop research activities and systematic interactions in areas of mathematical analysis and statistics directly connected to signal and image processing. While standard research is carried out at an international level at all participating institutions in cooperation with their individual partners the proposal is going for integration at a higher level. Thus it is one of the declared goals of HASSIP to shorten the development cycle for new algorithms by bringing together those who are involved in this process: the mathematicians and physicists working on the foundations (with view towards applications), the partners doing applied research (mostly engineering departments), are more experienced when it comes to implementations. High tech industry worldwide is in need for know-how and trained people in this area, and has shown very strong interest in our programme. The main research goal is therefore to improve the link between the foundations and real word applications, by developing new nonstandard algorithms, by studying their behaviour on concrete tasks, and to look for innovative ways to circumvent shortcomings or satisfy additional request arising from the applications. In doing so we expect to stimulate research in this area and the interactions in theoretical and applied domain.

From the mathematical point of view, the main areas involved are:

  • Harmonic analysis, in particular time-frequency and time-scale analysis.
  • Statistics, with a particular emphasis on statistical models for signals and images.
  • Numerical analysis for signal processing applications.

The research activities of the network will be centered around a few selected applications:

  • Audio signal modeling and compression.
  • Image and video encoding and compression.
  • Signal and image restoration and modification.

Partners

Contributions of the University of Marburg

The Marburg group is primarily concerned with the development of efficient wavelet and frame algorithms.
  • One central research topic is the construction of wavelets/multiwavelets for general scalings and of general Banach frames for coorbit spaces.
  • Moreover, we are interested in associated approximation schemes such as nonlinear and adaptive algorithms.
  • Our work also covers the development of efficient adaptive numerical schemes for operator equations.
  • Within the HASSIP network, we shall especially continue our recent work on generalized coorbit spaces on manifolds. The construction of suitable Banach frames and the development of associated numerical algorithms will be one of our main research topics.
We intend to collaborate in particular with NuHAG in Vienna, with ZeTeM in Bremen and with the Marseille team.

Workshop:

We are happy to announce the workshop "Recent Progress in Wavelet Analysis and Frame Theory" taking place from 23 to 26 January 2006 in Bremen, Germany. It will be a joint event of HASSIP and the DFG priority program 1114 and will be organized by the Bremen and Marburg groups.