Fecha: 10/05/2013 13:00
Lugar: Seminario Heaviside, Edificio de Tecnologías de la Información y las Telecomunicaciones, 2ª planta.
Grupo: GIR Análisis Numérico de Problemas de Evolución
Abstract:
In this talk, a remote sensing technique that infers the three-dimensional wave
form and radiance of oceanic sea states via a variational stereo imagery formulation is presented. This technique combines elements from variational optimization, image processing, computer vision, differential geometry, PDEs, numerical methods, multigrid methods and ocean engineering. The problem is formulated so that the shape and radiance of the wave surface are minimizers of a composite cost functional which combines a data fidelity term and smoothness priors on the unknowns. The desired ocean surface shape and radiance are the solution of a system of coupled partial differential equations derived from the
optimality conditions of the cost functional. The proposed method is extended to study the space-time dynamics of ocean waves. Statistical and spectral analysis are carried out to validate the technique from a physically justified point of view. The results shows evidence of the fact that the omni-directional wavenumber spectrum S(k) of the reconstructed waves decays as Zakharov’s theory predicts. The three-dimensional spectrum of the reconstructed wave surface is exploited to estimate wave dispersion and currents.