FLUid Image analysis and Description

This interdisciplinary project, that we have entitled FLUID, aims at studying and developing new methods for the estimation, the analysis and the description of complex fluid flows from image sequences. We propose to devise novel image processing and computer vision methods by using sound methodological frameworks incorporating suitable physical models accounting for the observed phenomena.This domain of research encompasses a wide range of difficult issues and thus will have a significant impact on several scientific and application domains including meteorology, oceanography, and flow visualization in applied fluid mechanics.The first objective of our proposal consists in studying novel and efficient methods to estimate and analyse fluid motions from image sequences. The second objective is to guarantee the applicability of the developed techniques to a large range of fluid visualization applications. To that end, two specific areas will be considered: meteorological applications and experimental fluid mechanics for industrial evaluation and control. From the application point of view the project will particularly focus on 2D and 3D wind field estimation, and on 2D and 3D particle image velocimetry. A reliable description of the fluid flow velocity will allow us to study techniques for the tracking of turbulent structures in the flows. Such problem is very important in meteorology to monitor or to predict automatically displacements of tornadoes and vortices.

Contact: Etienne.Memin @ irisa.fr                   
Phone :              +33