Data fusion applications: classification and mapping
Sophie Fabre, Pierre Dhérété · 2004
The nonprobabilistic theories have proved in the last years their ability to solve a large range of problems concerning imprecision. More recently, techniques using the Dempster-Shafer's and fuzzy set theories tried to deal with the problem related to the management of the uncertainty, the imprecision and the data fusion. The main difficulty of these methods concerns the knowledge modeling. We present two fusion applications: classification and mapping using both Dempster-Shafer's and fuzzy set theories in order to combine heterogeneous information. These techniques are proposed to improve multispectral classifications, GIS updating and mosaic building.