Dynamic estimation of the discernment frame in belief function theory

Belief function theory is now widely used in decision systems because of its ability to model both the imprecision and the uncertainty. Now, the problem of the discernment frame estimation is even more critical as the whole set of handled hypotheses is the powerset of the discernment frame. In this work, we propose to update and adjust the discernment frame in a sequential way as new sources provide new information pieces. Besides incompleteness, we assume that the current discernment frame may contain duplicated or fictitious hypotheses. We thus propose new updating mechanisms and we show on a practical application, namely the video surveillance, how these mechanisms work.