Observatoire de Paris Institut national de recherche scientifique français Univerité Pierre et Marie Curie Université Paris Diderot - Paris 7


  • Jeudi 29 septembre 2022 à 16h00 (Salle de réunion du bâtiment 16 et visioconférence)

    Data reduction and inverse problem approaches - An efficient (and cheap) way to push the instrumental limits

    Anthony BERDEU (LESIA)

    Study of astrophysical objects requires always more complex models (black hole surrounding, exoplanet atmosphere, accretion disk, galaxy formation and evolution, …) whose inputs imply always more precise measurements. On the other side, the astronomical instruments start to reach their fundamental limits (negligible sensor readout noise, precise mirror polishing, technologies harder to scale up to extremely large telescopes, …). During my talk, I will discuss how data science applied for optimal data reduction and processing via inverse problem approaches can bridge this gap, by pushing the experimental limits without the need of further instrumental developments. I will introduce different applications of my research : integral field spectroscopy (SPHERE/IFS), blind deconvolution and PSF reconstruction (SPHERE/ZIMPOL) and extreme adaptative optics for coronagraph (Evanescent Wave Coronagraph, EvWaCo).