Volume 387, Number 1, May III 2002
|Page(s)||347 - 355|
|Section||Numerical methods and codes|
|Published online||15 May 2002|
Optimal extraction of multiple overlapping spectra using a maximum entropy algorithm
Institut d'Astrophysique et de Géophysique, Université de Liège, avenue de Cointe 5, 4000 Liège, Belgium
2 Astronomical Observatory of Kyiv University, Observatorna St. 3, UA-04053 Kyiv, Ukraine e-mail: firstname.lastname@example.org
Corresponding author: J. Surdej, email@example.com
Accepted: 30 January 2002
A general algorithm of slit spectra extraction for a system of point-like sources (e.g. multiple lensed images of a quasar) has been developed, assuming that the point-spread function (PSF) induced by the measuring instrument and/or atmosphere and the positions of the spectra relative to the CCD frame are unknown. The main idea of the algorithm is to successively apply the maximum entropy method to each set of parameters, such as the spectra, the PSF, and the spectra positions, in order to iteratively improve their values. The algorithm uses all the a priori knowledge about the spectra (e.g. flux positivity, flux ratios between the components, astrometry, etc.) to compute the initial parameter sets. The main features of the algorithm, its implementation, as well as some important aspects of its practical use, are discussed in detail. Two sets of simulated spectroscopic data have been built in order to show the most characteristic properties of the algorithm and to justify its aplication to the spectra extraction of the gravitational lens system Q1009-0252 A & B (= LBQS1009-0252 A & B). Further applications of the algorithm are suggested.
Key words: techniques: image processing / techniques: spectroscopic
© ESO, 2002
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.