Tools and methods used in data fusion

Data fusion provides the means to combine disparate but complementary data, giving a more complete vision of the characteristics of the objects observed. This paper intends to present an overview on tools and methods existing in the field of data fusion for Earth observation data exploitation. Starting with a general introduction to the concept the article concentrates on the description of techniques to fuse data. The methods explained are categorised according to their processing level at which the fusion itself is performed. Although the paper aims at giving a global overview special attention is paid to pixel based techniques with regards to the fact that remote sensing data is provided in raster format. At the end an example of successfully implemented data fusion is illustrated by results of a study performed at the Western European Union Satellite Centre. Data fusion is used to improve the visual image exploitation by human image interpreters.