Linear Regression Models and Neural Networks for the Fast Emulation of a Molecular Absorption Code

Guillaume Euvrard (1) , Isabelle Rivals (1) , Thierry Huet (2) , Sidonie Lefebvre (2) , Pierre Simoneau (2)

1 ESA - Equipe de Statistique Appliquée (UMRS 1158)
2 ONERA - The French Aerospace Lab [Châtillon]

The background scene generator MATISSE, whose main functionality is to generate natural background radiance images, makes use of the so-called Correlated K (CK) model. It necessitates either to load or to compute thousands of CK coefficients for each atmospheric profile. When the CK coefficients cannot be loaded, the computation time becomes prohibitive. The idea developed in this paper is to substitute fast approximate models to the exact CK generator: using the latter, a representative set of numerical examples is built and used to train linear or nonlinear regression models. The resulting models enable an accurate CK coefficient computation for all the profiles of an image in a reasonable time.

Keywords: Numerical approximation and analysis; Absorption; Radiative transfer; Neural networks; Atmospheric propagation 110.2960 Image analysis

Domains: Neural and Evolutionary Computing [cs.NE]; Numerical Analysis [cs.NA]; Signal and Image processing; Signal and Image Processing

Guillaume Euvrard, Isabelle Rivals, Thierry Huet, Sidonie Lefebvre, Pierre Simoneau. Linear Regression Models and Neural Networks for the Fast Emulation of a Molecular Absorption Code. Applied optics, 2009, 48 (35), pp.6770-6780. ⟨hal-00805086⟩