Skip to Main content Skip to Navigation

Modélisation physique d'images de télédétection optique

Abstract : In optical remote sensing, radiative transfer models applied to Earth landscape and atmosphere allow one to simulate and analyze satellite measurements. Used in direct mode, these models are valuable tools for specifying technical characteristics of future satellite sensors. Used in inverse mode, they allow one to estimate surface parameters from satellite measurements acquired with any observation configuration and with few ground data. Presently, the major limitation of radiative transfer models comes from the strong simplifying hypotheses used to represent the geometry of landscapes (especially for vegetation). This constraint affects the radiometry and texture of remote sensing images at any spatial resolution. It is particularly evident for very high spatial resolution images. This stresses the interest of models that simulate images while using a realistic 3-D representation of Earth landscapes. The major objective of this thesis was to improve the accuracy and robustness of a 3D radiative transfer model called DART. The model simulates radiation propagation with ray tracing and discrete ordinate methods within a medium composed of turbid cells (for vegetation and air) and cells containing opaque figures (for walls, soil, trunks, etc.). The accuracy was improved with the introduction of new simplifying hypotheses concerning radiative transfer modeling applied to turbid and opaque cells. The robustness was greatly improved with the direct modeling of radiative transfer in the atmosphere and with the possibility to simulate any kind of landscapes (urban or natural). DART accuracy was validated comparing it to other radiative transfer models and to airborne images of an urban district. The model proved to be efficient because it was used to specify the technical characteristics of the future very high spatial resolution sensor of the satellite constellation Plé iades. DART was also applied to the estimation of biophysical parameters (foliar index, crown coverage and chlorophyll concentration) of a temperate forest using SPOT (20 meters resolution) and Ikonos (1 meter resolution) satellite images.
Complete list of metadatas

Cited literature [86 references]  Display  Hide  Download
Contributor : Ferran Gascon <>
Submitted on : Sunday, July 31, 2005 - 11:04:19 AM
Last modification on : Thursday, October 15, 2020 - 3:18:36 AM
Long-term archiving on: : Friday, November 25, 2016 - 9:21:47 AM


  • HAL Id : tel-00009863, version 1



Ferran Gascon. Modélisation physique d'images de télédétection optique. Physique [physics]. Université Paul Sabatier - Toulouse III, 2001. Français. ⟨tel-00009863⟩



Record views


Files downloads