Search

All raw data recorded by the Very Large Array (VLA) since it began operation in 1980 have been archived, test observations and calibrations as well as program observations. The reels of tape are kept in a vault in the control building of the VLA on the Plains of San Augustin in New Mexico; the vault was filled several years ago and so recordings for recent years are in a trailer parked outside the control building. The data are in the public domain, by policy, with a proprietary period of 18 months for program observations. A subset of the observation catalog (all non-calibrator observations with duration longer than one hour) is distributed as a set of floppy disks (MS-DOS format) with a search program (for MS-DOS); the full catalog is available on a computer at the Array Operations Center in Socorro, New Mexico. One technical problem is that the existing catalog doesn’t tabulate which reel of tape in the vault has which day’s data. Reel numbers must be obtained from a log book at the VLA site. But the data tapes can be located in the vault, they are coded in a modest set of reasonably self-documenting formats, and a portable program which is exported by the National Radio Astronomy Observatory (NRAO) in all AIPS (Astronomical Image Processing System) kits can be used to read files copied from them.

Over the last few decades, astronomical image processing has become extremely sophisticated, encompassing image reconstruction and restoration, image filtering, object detection and classification. A collaboration from the Space Telescope Science Institute, in Baltimore; Johns Hopkins University; and the Lombardi Cancer Research Center at the Georgetown University Medical Center, in Washington, DC, is hoping to apply some of these methods to detect telltale signs of breast cancer in a digitized mammogram. The project was catalyzed over a year ago by Benjamin Snavely, program director for advanced technologies and instrumentation in the NSF division of astronomical sciences, which recently awarded the collaboration a $50 000 grant from the Small Grants for Exploratory Research Program.

Astronomical Image Processing (AIP) needs extremely interactive software systems. It is therefore necessary to execute in a short time frequently used, time consuming elaborations, such as convolution, geometrical normalization and geometrical correction. In this paper such operations have been splitted in a scalar part (typical of the operation) and in a vectorial part (in common among the whole operation set). In this way it is possible to achieve their optimization with regard to a native or attached vector processor. The common vectorial part (vectorial nucleus) is studied in detail and checked on VAX 11/780+FPS 5410, CRAY X-MP/12 and IBM 3090VF processors.

This reply is prompted by Fred Maryanski's question in his Chairman's column (Sigsmall Newsletter, August 1982). "What will we do with all that bandwidth?" This response follows my consideration of the implications for on-line video signal processing given the advent of 2Ø MHz bus, 8 Mbyte, 32-bit minicomputer systems. This is what "think Mini" now means for us in astronomical image processing applications. We are now considering the question: What happens in a modestly networked video signal processing environment?

The topics dealt with in this paper are as follows. Firstly, to set the scene, we briefly overview the major astronomical image processing systems, and make reference to current software engineering problems in this area. Secondly, we survey a range of pattern recognition problems, which all have classification as their central objective. These pattern recognition problems are: object searching and classification in photometry; the classifying of galaxies on the basis of their morphological shapes; and the classification of stellar spectra. Finally, we review some of the specific problems expected for Hubble Space Telescope image data.

Multiresolution transforms have been used with great effectiveness in many areas of astronomical image processing. Here we seek to extend their applicability in two directions. Firstly, we use such transforms to detect objects in fields containing point sources, or a mixture of point sources and large, extended objects. Secondly, we discuss the efficient transfer of (compressed) images using current World-Wide Web technology.

We discuss a collection of programs to display and analyze astronomical images on IBM-PC and compatible computers. These programs, based on the VISTA program at Lick Observatory and on DAOPHOT at Dominion Astrophysical Observatory, have facilities to display data in FITS format on a variety of monitors, to perform elementary arithmetic operations on images, to do simple photometry, and to extract one-dimensional information. The merits of PC versus minicomputer-based image processing are discussed.