Volume 565, May 2014
|Number of page(s)||10|
|Published online||05 May 2014|
An image reconstruction method (IRBis) for optical/infrared interferometry
Max-Planck-Institut für Radioastronomie, Auf dem Hügel 69, 53121 Bonn, Germany
Received: 11 December 2013
Accepted: 25 February 2014
Aims. We present an image reconstruction method for optical/infrared long-baseline interferometry called IRBis (image reconstruction software using the bispectrum). We describe the theory and present applications to computer-simulated interferograms.
Methods. The IRBis method can reconstruct an image from measured visibilities and closure phases. The applied optimization routine ASA_CG is based on conjugate gradients. The method allows the user to implement different regularizers, apply residual ratios as an additional metric for goodness-of-fit, and use previous iteration results as a prior to force convergence.
Results. We present the theory of the IRBis method and several applications of the method to computer-simulated interferograms. The image reconstruction results show the dependence of the reconstructed image on the noise in the interferograms (e.g., for ten electron read-out noise and 139 to 1219 detected photons per interferogram), the regularization method, the angular resolution, and the reconstruction parameters applied. Furthermore, we present the IRBis reconstructions submitted to the interferometric imaging beauty contest 2012 initiated by the IAU Working Group on Optical/IR Interferometry and describe the performed data processing steps.
Key words: instrumentation: interferometers / methods: data analysis / methods: numerical / techniques: interferometric / techniques: image processing / techniques: high angular resolution
© ESO, 2014
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.