Back to overview

Tensor optimization for optical-interferometric imaging

Type of publication Peer-reviewed
Publikationsform Original article (peer-reviewed)
Publication date 2014
Author Auria Anna, Carrillo Rafael, Thiran Jean-Philippe, Wiaux Yves,
Project Advanced signal processing on the sphere for high angular resolution diffusion magnetic resonance imaging
Show all

Original article (peer-reviewed)

Journal Monthly Notices Of The Royal Astronomical Society
Volume (Issue) 437(3)
Page(s) 2083 - 2091
Title of proceedings Monthly Notices Of The Royal Astronomical Society
DOI 10.1093/mnras/stt1994

Open Access

Type of Open Access Repository (Green Open Access)


Image recovery in optical interferometry is an ill-posed non-linear inverse problem arising from incomplete power spectrum and bispectrum measurements. We reformulate this non-linear problem as a linear problem for the supersymmetric rank-1 order-3 tensor formed by the tensor product of the vector representing the image under scrutiny with itself. On one hand, we propose a linear convex approach for tensor recovery with built-in supersymmetry, and regularizing the inverse problem through a nuclear norm relaxation of a low-rank constraint. On the other hand, we also study a non-linear non-convex approach with a built-in rank-1 constraint but where supersymmetry is relaxed, formulating the problem for the tensor product of three vectors. In this second approach, only linear convex minimization subproblems are, however, solved, alternately and iteratively for the three vectors. We provide a comparative analysis of these two novel approaches through numerical simulations on small-size images.