Multisource classification using ICM and Dempster-Shafer theory
Samuel Foucher, Mickaël Germain, J.-M. Boucher, G.B. Bénié · IEEE Transactions on Instrumentation and Measurement · 2002
We propose to use evidential reasoning in order to relax Bayesian decisions given by a Markovian classification algorithm, the multiscale iterated conditional mode (ICM) algorithm. The Dempster-Shafer rule of combination enables us to fuse decisions in a local spatial neighborhood which we further extend to be multisource. This approach enables us to more directly fuse information. Application to the classification of very noisy images produces interesting results.