ISO SOC Team.
Next: IMAGER: A Parallel Interface to Spectral Line Processing
Previous: Calibration with the ISOPHOT Interactive Analysis (PIA)
Up: Science Software Applications
Table of Contents - Index - PS reprint
C. Gabriel1
European Space Agency - Astrophysics Division
I. Heinrichsen,1 D. Skaley2
Max-Planck Institut für Kernphysik, Heidelberg, Germany
W.-M. Tai
Dublin Institute for Advanced Studies, Dublin, Ireland
1Currently at the ISO Science Operations Centre, Villafranca, Apdo 50727, E-28080 Madrid, Spain
2Now at the Max-Planck Institut für Radioastronomie, Bonn, Germany
This article is devoted to a description of the image processing capabilities of PIA, on the basis of the different mapping strategies with ISOPHOT. PIA offers a full graphical interface, giving the user all the informations related to the observation and data selection possibilities. Special flat fielding techniques, extraction of profiles, map rotation and convolution, point source extraction, three dimensional display, etc., are implemented in an interactive way.
The ISO PHOT Interactive Analysis PIA is described elsewhere in this volume (Gabriel et al. 1997). It was conceived primarily as a calibration tool for ISOPHOT (Lemke et al. 1996), one of the four instruments on board ISO, the Infrared Space Observatory (Kessler et al. 1996). However, the software package has been developed into a full interactive astronomical analysis system for ISOPHOT data.
ISOPHOT performs infrared mapping in different modes using different detector subsystems. A detailed description of these modes can be found in the ISOPHOT Observer's Manual (Klaas et al. 1994). Mapping can be performed using one of the three single pixel photometers (P1, P2, P3) in the range 2.5-100µm, or one of the two far infrared cameras (C100, a 3×3, and C200, a 2×2 detector array) in the range 50-240µm. Several filters in those ranges and apertures (in the case of the single photometers) are available.
The raster capability of ISO, pointing sequentially to several positions on a two-dimensional grid, makes it possible to have the combination of individual fluxes and an image of a sky region. ISOPHOT measures continuously with a fixed instrument configuration during the raster performance. The PIA mapping software basically combines the sky brightnesses measured at the different positions to an image in sky coordinates.
A raster observation can have a maximum of 32×32 raster points with a maximum sampling area of 1.6×1.6°. The ISO Observer's Manual (ISOOBS 1994) gives full information on different aspects of ISO's raster observations.
There are two mapping modes possible using ISOPHOT:
As input for mapping processing, PIA uses ISOPHOT data which has been reduced to the level of sky brightness (in MJy/sr) for each raster (and chopper) step and per detector pixel, together with the associated pointing information per raster point. All these data, corresponding to one measurement, are contained in an element of the so called AAP (astrophysical application) data structure. A description of the data reduction from the raw telemetry to this level as done by PIA can be found in Gabriel et al. (1996).
On this level, we must deal with data derived from a measurement: an array of measured brightnesses, their uncertainties, associated sky positions and corresponding observation times of these positions. PIA calculates the positions of the individual detector pixels at different raster/chopper positions during the raster measurement. After this calculation, the values of the detector signals are binned into map pixels. A simple gridding function is applied, which is a trapezoidal function, i.e., the geometric overlap of detector pixel and map pixel is used as the contributing part of the measured detector brightness to the map pixel. For the final image computation, PIA uses the coverage and the time of coverage of all the contributing signals for normalization.
PIA produces three kinds of maps for each measurement:
PIA allows a free choice for the image binning, although there is a ``natural'' choice, given by the level of oversampling reached in the observation. The pointing taken for the image computation is given by the measured positions, which can differ slightly from the planned ones.
For data obtained with one of the detector arrays, there is also the possibility of selecting/deselecting detector pixels to use only part of the data, or obtaining maps from individual detector pixels. This may help judge the quality of parts of the map, reveal flat fielding problems, etc.
Flat fielding of the detector arrays is in principle given, since measurements of the internal fine calibration sources are performed before and after the raster source measurement. Nevertheless, the possibility of using an additional flat fielding technique is given. If this option is chosen, individual maps are produced for every detector pixel and the central, common region of the maps taken for obtaining flat fielding factors.
Once the map is produced, the PIA graphical interface offers several possibilities for enhancing the quality of the image display. Starting from a Map Display Window, several context sensitive menus allow changing the color tables, interpolating image pixels, zooming every map region with different zoom factors, setting cut values, overplotting contours to the map, obtaining profiles, extracting flux values and positions from the map, and extracting possible point sources.
It is also possible to obtain a three-dimensional surface from the map, using an interface allowing rotation about each axis and super-position of contours.
PIA includes the option of convolving a map to a given spatial resolution. To facilitate comparisons between maps obtained in different wavelengths, conversion to the resolution of every PHT-filter is available. A two-dimensional Gaussian approximation to the point spread function is used for the convolution.
Maps can be also rotated by PIA to every angle with respect to the RA-DEC plane.
The main input for mapping with PIA is the AAP data, resulting from the reduction of a raster measurement. All of the mapping capabilities just described may be applied to these data.
Several additional formats can be used both as input to and output from the PIA imaging software:
The very simple trapezoidal gridding function used for obtaining an image by PIA will be complemented with enhanced imaging methods. Using the redundant information from a single detector pixel should lead to better spatial resolution. The same redundant information will be used for modeling detector response transients during an observation. Flat-fielding may also be enhanced by taking a user defined area (preferably a background area) for computing the flat-fielding factors. Measured point spread profiles will be used for image convolution and for point source extraction.
FITS 1993, Definition of the Flexible Image Transport System (FITS), Standard, NOST 100-1.0, NASA / Science Office of Standards and Technology, Code 633.2, NASA Goddard Space Flight Center, Greenbelt, Maryland
ISOOBS 1994, ISO Observer's Manual, ISO SOC Team.
Gabriel, C., Acosta-Pulido, J., Heinrichsen, I., Morris, H., & Tai, W.-M. 1997, this volume
Gabriel, C., Haas, M., Heinrichsen, I., & Tai, W.-M. 1996, ISOPHOT Interactive Analysis User Manual3
Kessler, M. F., et al. 1996, A&A, 315, 27
Klaas, U., Krüger, H., Heinrichsen, I., Heske, A., & Laureijs, R. (eds), 1994, ISOPHOT Observer's Manual3
Laureijs, R., Richards, P. J., & Krüger, H. 1996, ISOPHOT Data Users Manual,3 V2.0
Lemke, D., et al. 1996, A&A, 315, 64
Next: IMAGER: A Parallel Interface to Spectral Line Processing
Previous: Calibration with the ISOPHOT Interactive Analysis (PIA)
Up: Science Software Applications
Table of Contents - Index - PS reprint