Land cover discrimination from multitemporal ERS images and multispectral Landsat images: a study case in an agricultural area in France

More and more remote sensing data corresponding to various wavelength domai ns is becoming available. Visible/infrared data were first used for land co ver classification. However, radar data are becoming more widely used for h ydrological and agricultural applications. This paper discusses the perform ance, for land cover type discrimination, of an optical image acquisition a nd a multitemporal radar series. For the majority of land cover types exist ing within the test site (representative of northern European agricultural areas), both ERS multitemporal SAR and Landsat multispectral visible/infrar ed classifications lead to good results, with the latter being more robust. For better identification of cultures that are less represented, the compl ementarity of the two datasets may be exploited using an efficient data fus ion algorithm based on the Dempster-Shafer evidence theory. The performance of this combination was verified on two successive vegetation cycles.