Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.adass.org/adass/proceedings/adass00/reprints/P1-12.pdf Дата изменения: Wed May 30 01:21:38 2001 Дата индексирования: Tue Oct 2 10:34:37 2012 Кодировка: Поисковые слова: arp 220

Astronomical Data Analysis Software and Systems X ASP Conference Series, Vol. 238, 2001 F. R. Harnden Jr., F. A. Primini, and H. E. Payne, eds.

AIRY: Astronomical Image Restoration in interferometrY
Serge Correia, Marcel Carbillet, Luca Fini Arcetri Astrophysical Observatory, Largo E. Fermi 5, 50125 Firenze, Italy Patrizia Boccacci, Mario Bertero INFM and Department of Computer and Information Sciences, University of Genova, Via Dodecaneso 35, 16146 Genova, Italy Antonella Vallenari Padova Astronomical Observatory, vicolo del l' Osservatorio 5, 35122 Padova, Italy Andrea Richichi European Southern Observatory, Karl-Schwarzschildstr. 2, 85748 Garching b.M., Germany Massimo Barbati Astronomical Observatory of Torino, Strada Osservatorio 20, 10025 Pino T.se (TO), Italy Abstract. AIRY is a modular software package designed to simulate optical and near-infrared interferometric observations and/or to perform subsequent image restoration/deconvolution. It is written in IDL and has been designed to be used together with the CAOS Application Builder, version 2.0 or higher. AIRY can be applied to a wide range of imaging problems. We will present in particular an application to the case of interferometric imaging with the Large Binocular Telescope, in which we simulate the observation and scientific interpretation of a synthetic star cluster in the near-infrared. (Related Web page1 )

1.

Introduction

AIRY is the acronym describing the activity of a group of astronomers and mathematicians from various Italian institutions (see the Web page indicated above). The aim of the collaboration is to develop methods and software for the restoration of interferometric images, with application to the Large Binocular Telescope (LBT). One of the first results is the package AIRY, IDL-based and
1

http://dirac.disi.unige.it

404 c Copyright 2001 Astronomical Society of the Pacific. All rights reserved.


AIRY: Astronomical Image Restoration in interferometrY

405

CAOS-compatible (Fini et al. 2001). AIRY is designed to simulate optical and near-infrared interferometric observations and/or to perform subsequent image restoration/deconvolution. It consists of a set of specific modules which are listed and briefly presented in Section 2. The package also includes a library of ideal and Adaptive Optics (AO) corrected LBT point-spread functions (PSFs). Details can be found in Carbillet et al. (2001). An interesting feature of AIRY is its multiple deconvolution capability, well suited for the LBT case. In the current version the method implemented is the so-called Ordered Subsets - Expectation Maximization (OS-EM) algorithm (Bertero & Boccacci 2000a). As an example of application we present in Section 3 a simulated LBT observation together with a scientific interpretation of the results. 2. The Modules of the AIRY Simulation Package

Table 1 shows a complete list, together with a very brief description, of the modules of the current version of the AIRY Simulation Package. Table 1. Module Data simulation modules OBJ - OBJect definition Modules of the AIRY Simulation Package. Purpose to define the ob ject characteristics among several ob ject types (binary ob ject, open cluster, planetary nebulae, SN remnant, spiral galaxy, YSOs, stellar surface, user-defined) to perform convolution to add the noise contributions to perform image pre-processing to perform deconvolution (Ordered Subsets-Expectation Maximization method) to analyse reconstructed images of binary ob jects to detect stars in the reconstructed images to read FITS images to write FITS images to restore image structure cubes (XDR format or FITS format) to save image structure cubes (XDR format or FITS format) to display images

CNV ADN Data PRE DEC -

object-PSF CoNVolution ADd Noise to image processing modules PRE-processing DEConvolution process

Data analysis modules ANB - ANalysis of Binary FSM - Find Star Module Othe RFT WFT RSC r modules and utilities Read FiTs file format Write FiTs file format Restore im. Struct. Cubes

SIM - Save IMage struct. DIS - DISplay image


406

Correia et al.

Figure 1.

Example of worksheet of a typical simulation.

Figure 1 shows an example of a simulation that can be built with the AIRY Simulation Package. This simulation is essentially composed of three parts. The first part models the observed data by convolving an ob ject map (here a stellar cluster) of given characteristics with a set of PSFs (ob ject-PSFs), extracted, for this example, from the library. The different noise contributions are then added. The second part is the restoration of the observed data set by multiple deconvolution with another set of PSFs (reference-PSFs), after a pre-processing stage. The third part permits the analysis and saving of the deconvolved image. Note that the modular structure of AIRY also allows using the package for improving real AO data by removing part of the AO-correction residual, and/or to produce images from real interferometric data. 3. Example Application: Scientific Analysis of a Simulated Star Cluster Observed with the Large Binocular Telescope

The goal is to simulate high-resolution interferometric observations of a scientific ob ject of interest with LBT, and to retrieve the scientific parameters of this ob ject after the image restoration process. We have considered a star cluster composed of 1898 stars with the following characteristics: age 4.0 Gyr, metallicity Z=0.008, distance modulus=19( 63 kpc), reddening=0, extension field= 10. 24 в 10. 24. Three ob ject maps (2048 в 2048 pixels) were modeled in J, H, and K bands. The resulting magnitude ranges were respectively 14.01­24.22, 13.25­23.63 and 12.89­23.56. The worksheet of this simulation is similar to that presented in Figure 1 for each band. We have simulated observations at three parallactic angles (0 , 60 , and 120 ) for each band, and with 2000 s integration time for each parallactic angle. PSFs were assumed ideal (coherence, cophasage, no aberrations) for both the reference and the ob ject-PSFs. Multiple deconvolution was carried out for each band using 100 iterations of the OS-EM algorithm (see Bertero & Boccacci 2000a, Bertero & Boccacci 2000b, Bertero et al. 2000). Detection and photometry on the restored frames were performed using DAOPHOT, with a 25-sigma detection threshold and a 3 pixels (15 mas) aperture photometry diameter.


AIRY: Astronomical Image Restoration in interferometrY

407

Figure 2. CMD of the star cluster in the plane H vs. (J-H). Detected stars are marked with squares and isochrones corresponding to different ages and for metallicity Z= 0.001 (left), Z= 0.008 (center), and Z= 0.02 (right) are superimposed. The Color-Magnitude Diagram (CMD) of the star cluster in the plane H vs. (J-H) is shown in Figure 2. Detected stars are marked with squares and isochrones corresponding to different ages and metallicity values are superimposed. From a first visual inspection, we can derive that Z 0.008 and the age is 4.0­5.0 Gyr, in good agreement with the input parameters. References Bertero, M., & Boccacci, P. 2000a, Bertero, M., & Boccacci, P. 2000b, Bertero, M., Boccacci, P., Correia, Optical Astronomy, ed. P. J Carbillet, M., Fini, L., Femenґ ia, Delplancke, F., & Hubin, N. Fini, L., Carbillet, M., & Riccardi, A&AS, 144, 181 A&AS, 147, 323 S., & Richichi, A. 2000, in Interferometry in Lena & A. Quirrenbach, SPIE 4006, 514 B., Esposito, S., Riccardi, A., Viard, E., 2001, this volume, 249 A. 2001, this volume, 253