Skip to Main content Skip to Navigation

Solving ill-posed Image Processing problems using Data Assimilation

Abstract : Data Assimilation is a mathematical framework used in environmental sciences to improve forecasts performed by meteorological, oceanographic or air quality simulation models. Data Assimilation techniques require the resolution of a system with three components: one describing the temporal evolution of a state vector, one coupling the observations to this state vector, and one defining the initial condition. In this report we use this framework to study a class of ill-posed Image Processing problems, usually solved by spatial and temporal regularization techniques. A generic approach is proposed to convert an ill-posed Image Processing problem in terms of a Data Assimilation system. This method is illustrated on the determination of optical flow from an image sequence. The main advantage of the resulting software is the use of a quality criteria on observations for weighting their contribution in the estimation process and of a dynamic model to ensure a relevant temporal regularity of the result.
Complete list of metadatas
Contributor : Dominique Béréziat <>
Submitted on : Tuesday, March 17, 2009 - 9:29:31 PM
Last modification on : Friday, August 31, 2018 - 9:25:53 AM
Document(s) archivé(s) le : Tuesday, June 8, 2010 - 9:37:44 PM


Files produced by the author(s)


  • HAL Id : inria-00368890, version 1


Dominique Béréziat, Isabelle Herlin. Solving ill-posed Image Processing problems using Data Assimilation. [Research Report] 2009, pp.33. ⟨inria-00368890v1⟩



Record views


Files downloads