Scenario recognition in medical environment using temporal HMM integrating Transferable Belief Model (TBM)

and Motivation In recent years, Multi-sensors based-tracking and recognition become a major problems for the computer vision community. This work addresses the problem of recognizing human activities in medical Units Care Intensives, which is an important research issue for building a medical pervasive and smart environment. For this purpose, we propose an approach based on a temporal HMMs integrating explicitly the sate duration using Transferable belief model reasoning. Time and again hidden Markov models have been demonstrated to be highly effective in one-dimensional pattern recognition and classification problems such as speech recognition. It is beneficial to exploit the explicit state duration for temporal scenario recognition because the temporal parameters are very important for activity recognition. Meanwhile, the TBM-based modeling of the multi-visual-sensors fusion allows to take into account imprecision, uncertainty, conflict inherent to the features and incomplete data. Significant experiments were carried out on video sequences taken from video-surveillance system in hospital's Heart Section. . 1 0 O p j t t p o b a i j x N j i i t j s j t t s ij t t t        