Soup.GroundPlane.org

“ intelligent data fusion is the combination of unbiased pattern recognition techniques with expert information about the strengths and weaknesses associated with the data acquisition techniques and expert knowledge of the geology, foliage, and man-made features located at the site under scrutiny. Data from multiple sensing modalities are utilized in an overall decision-making algorithm that is theoretically more accurate than any individual sensor on its own.

The two most well-known theoretical data fusion approaches are those formulated by Bayesian inference theory and by Dempster-Shafer theory. Bayesian theory, the older and more established of the two, is founded on a rigorous statistical framework and generally requires significant statistical assumptions regarding the input data. Dempster- Shafer can be thought of as a relaxation of Bayesian theory in which subject belief assessments are utilized and can be assigned to supersets of hypotheses as well as to single hypotheses.  ”