CCMA/ESF: The European Scientific Network on Converging Computing Methodologies in Astronomy

M. C. Maccarone

Ist. IFCAI/CNR, Piazza G. Verdi 6, 90138 Palermo, Italy

F. Murtagh

ST-ECF, ESO, Karl-Schwarzschild-Str. 2, 85748 Garching bei München, Germany

A. Bijaoui

Observatoire de Côte d'Azur, B. P. 229, 06304 Nice, France

Abstract:

1. Introduction

In January 1995 a new Scientific Network, supported by the European Science Foundation ESF was defined. The Network, entitled CCMA (Converging Computing Methodologies in Astronomy), received funding to be used over a period of three years for organizing workshops and work visits in Europe.

The motivation for this project derives from the fact that modern computational astronomy is characterized by the increasing integration of various methodologies and techniques in pattern recognition, information retrieval, and data analysis. Large astronomical databases from space-borne missions, ground-based observatories, and large wide-field surveys, provide the technological infrastructure.

European research has been always innovative and productive in the field of computational astronomy. European-based teams were among the first to carry out assessments in astronomy of wavelets, neural networks, and fractals, and to advance theory and practice of these methods.

In spite of this, no real coordination is currently available from any of the European organizations involved in astronomy or computing methodologies: the potential for having a cohesion and global view is only raised in a limited and partial way by various current organizational activities.

The main goal of the CCMA Scientific Network is then to make progress in the area of modern computational astronomy by rational study of the problems and their methodological solutions. We wish to highlight and exploit convergence of methods and techniques for treatment of information in the astronomical application domain; we are seeking harmonization and co-ordination, together with comparative advantages which accrue from this.

The CCMA Co-ordination Committee is composed of:

A. Bijaoui (Vice-chair), Obs. Côte d'Azur, Nice, France;
V. Di Gesú, Dept. Mathematics and Applic., University of Palermo, Italy;
A. Heck, Observatoire de Strasbourg, Strasbourg, France;
M. J. Kurtz, Harvard-Smithsonian CfA, Cambridge, MA, USA;
P. Linde, Lund Observatory, Lund, Sweden;
M. C. Maccarone (Chair), IFCAI/CNR, Palermo, Italy;
R. McMahon, Inst. of Astronomy, The Observatories, Cambridge, UK;
R. Molina, Dept. Computer Science and AI, University of Granada, Spain;
F. Murtagh (Secretary), ST/ECF at ESO, Garching, Germany;
E. Raimond, NFRA, Dwingeloo, The Netherlands;
and P. Colyer, ESF European Science Foundation, Strasbourg, France.

2. CCMA Network Activity

A series of three workshops will be organized during the three years of the CCMA. They will be oriented towards exchanging views, analyzing the situation, and producing recommendations. The lectures given at each workshop will be distributed in electronic form via the CCMA Web area and in printed form as a special issue of a scientific journal. The CCMA workshops will cover the following central topics:

• From vision models to image information retrieval. Methods such as wavelets, mathematical morphology, multiresolution approaches, and fuzzy sets have proven their worth in image data processing and analysis, and consequently in the framework of accessing appropriate information from large image databases. In this context, the methods must be moulded together to allow semantically-driven access to data. As an example, wavelet transform analysis is used for image restoration and filtering, with semantics built in through morphological operations on the transform. New methods, motivated by those mentioned, are required to handle the huge quantities of data which stand ready to be analyzed.

• The data life-cycle: methodological aspects. The astronomical data life-cycle is highly digital: data are subject to image processing and statistical treatment and the data characteristics strongly impact on methods for analysis and treatment. The final major stage in this process involves data archiving and publication. The issues of electronic publishing and of digital libraries are therefore viewed as increasingly important and data storage, access and interface standards require coordination.

• From data integration to information integration. Particular data integration or data fusion problems, such as integration of data associated with different wavelength ranges, are of great relevance in the context of large space- and ground-based observing projects. In certain instances, this takes the form of co-addition in image restoration. It also includes the enhancement of image restoration and filtering approaches through building-in semantic information on the cosmic objects of interest. Close, complementary use of multi-million object astronomical catalogs still has some distance to go in order to be tightly coupled with analysis of images from large image databases. Complementarity of such data rests on external information, and determines the appropriate analysis methodologies. Large-scale data integration prototypes pursued by NASA and ESA in recent years serve to point to ``issues and concerns'' which are solvable through improved methods. Moreover, compression is inherent to the integration process; it is a byproduct of image models and the loss of information involved in compression raises a range of questions. In its broadest sense, compression is summarization and therefore it is part of the overall process of scientific analysis.

A Conference on the Network's overall theme will be held shortly before the end of the Network, with the aim of bringing the results obtained to a wide audience. A selected set of papers will be published in book form.

3. First CCMA Workshop

The first CCMA workshop has been held in Nice, France, Observatoire de la Côte d'Azur, October 4--6, 1995. The workshop, entitled ``Vision Modeling and Information Coding'' and organized by A. Bijaoui, dealt with the nature of the information which can be obtained, and its computing, statistics, and pattern recognition expression. Such issues cannot be separated from the analysis methodologies, nor from the astronomical semantic context. A total of 23 lectures, given by European, USA and Australian researchers, was subdivided into three sessions:

• Vision: innovative modeling techniques in computer vision; possibilities and limitations of probabilistic models; concepts and applications of the symmetry transform; human vision model and wavelets for multi-resolution image coding; Hough transform and its application in astronomy; multiscale vision model applied to astronomical images; extracting, selecting and visualizing information from astronomical radio spectrograms; automated analysis of images and spectra for redshift surveys; challenges in crowded field astronomy; automatic construction of a catalog of galaxies from CCD images; image reconstruction; fuzzy mathematical morphology; vision model, wavelet transform and adaptive filtering.

• Restoration: image models and parameter estimation in astronomical image restoration; image reconstruction with variable resolution using visual response functions; image restoration with edge-preserving regularization; multiscale maximum-entropy image restoration; adaptive filtering and masking of HI data cubes.

• Compression: compression in astronomical data; compression by mathematical morphology; compression by wavelet transform; pyramidal image compression and object detection; classification in astronomical data.

All papers will be available on-line during 1995. The proceedings will be published as a special issue of a specialized scientific journal; an announcement will be made in the CCMA Web area.

The second CCMA workshop is planned for June 1996. It will be organized by A. Heck on the theme ``Strategies and Techniques of Information in Astronomy''. It will mainly address the information management problem; textual, bibliographic, graphical and image data in their dynamic aspects: production, selection, storage, processing, maintenance, retrieval, integration.