Chiffres-clé
Chiffres clefs
189 personnes travaillent au LJLL
86 permanents
80 chercheurs et enseignants-chercheurs permanents
6 ingénieurs, techniciens et personnels administratifs
103 personnels non permanents
74 doctorants
15 post-doc et ATER
14 émérites et collaborateurs bénévoles
Chiffres janvier 2022
Séminaire du LJLL - 07 12 2018 14h00 : G. Papanicolaou
George Papanicolaou (Université de Stanford)
Imaging sparse reflectivities from noisy data
Résumé
Algorithms for obtaining high-resolution images often use thresholding, which removes noise and other imperfections in the image efficiently provided that the support of the image is sparse and noise contamination is not too big. However, such imaging methods tend to be unstable since, above a certain level, the noise destroys the image. How can these imaging methods be stabilized ? I will review these issues and then present a method that in some cases can produce clean images even with a lot of noise. Numerical simulations illustrate the effectiveness of this approach.