Estimation du mouvement par assimilation de données dans des modèles dynamiques d'ordre réduit. (Motion estimation by data assimilation in reduced dynamic models)
نویسنده
چکیده
This paper concerns the estimation of apparent motion from an image sequence. A model, named full model, is defined to describe the dynamics of motion field and images. The Proper Orthogonal Decomposition (POD) is used to derive basis to characterize motion fields and images. The Galerkin projection of the full model on these basis defines the so-called reduced model. Coefficients of motion fields are retrieved from those of observations with the help of a 4D-var assimilation method. Last motion results are obtained from the coefficients. The method has been tested on twin experiments to validate the approach.
منابع مشابه
Strategies for processing images with 4D-Var data assimilation methods
Data Assimilation is a well-known mathematical technic used, in environmental sciences, to improve, thanks to observation data, the forecasts obtained by meteorological, oceanographic or air quality simulation models. It aims to solve the evolution equations, describing the dynamics of the state variables, and an observation equation, linking at each space-time location the state vector and the...
متن کاملUsing models of dynamics for large displacement estimation on noisy acquisitions
The paper discusses the issue of motion estimation on noisy images displaying large displacements, due to high velocity values. “Noisy” means that the data contain either missing acquisitions on isolated points, regions, frames or noisy measures. Assuming the dynamics is partially accessible from heuristics and modeled, the objective is to include this knowledge in the computation of the soluti...
متن کاملSolving ill-posed Image Processing problems using Data Assimilation. Application to optical flow
Data Assimilation is a methodological framework used in environmental sciences to perform forecasts with complex systems such as meteorological, oceanographic and air quality models. Data Assimilation requires the resolution of a system with three components: one describing the temporal evolution of the state vector, one coupling the observations and the state vector, and one defining the initi...
متن کاملPressure image assimilation for atmospheric motion estimation
The complexity of dynamical laws governing 3D atmospheric flows associated with incomplete and noisy observations makes the recovery of atmospheric dynamics from satellite images sequences very difficult. In this report, we face the challenging problem of estimating physical sound and timeconsistent horizontal motion fields at various atmospheric depths for a whole image sequence. Based on a ve...
متن کاملTowards a bio-inspired evaluation methodology for motion estimation models
Offering proper evaluation methodology is essential to continue progress in modelling neural mechanisms in vision/visual information processing. Currently, evaluation of motion estimation models lacks a proper methodology for comparing their performance against the visual system. Here, we set the basis for such a new benchmark methodology which is based on human visual performance as measured i...
متن کامل