Abstract de la publi numéro 4304

The Bidirectional Reflectance Distribution Function (BRDF) is certainly one of the most important surface properties used in remote sensing, for instance as input data of scene modelling. However, the use of measured BRDF is highly limited by its intrinsic complexity, and models that provide compression and additional functionalities are needed. One common way consists in fitting an analytical model to the set of measurements using an optimization technique. But this approach is always restricted to a specific class of surfaces, to a small angular or spectral range, and the modelling quality may strongly depends on the numerical algorithm chosen. Moreover, analytical models are unable to actually handle the BRDF dependence on wavelength. In this paper we present a new numerical model for acquired spectral BRDF to overcome these drawbacks. This model is based on a separation between the spectral and the geometrical aspect of BRDF, each projected into the appropriate wavelet space. After a brief introduction to BRDF, the advantages of wavelets and the construction of the model are explained. Then, results of modelling are presented over a large collection of real and simulated measurement data sets. At last, the robustness of the model is tested according to simulated noisy data.