Advances in Background Updating and Shadow Removing for Motion Detection Algorithms

摘要

The problem of detecting moving objects is very important in many application contexts such as people detection and recognition, visual surveillance both in indoor and outdoor environments, and so on. In this paper we propose two additional modules for a generic motion detection algorithm. The first one regards the background updating procedure: the novelty of the proposed algorithm is its capability, unlike traditional similar algorithms, to efficiently update each point of the reference model, even if covered by a foreground object. The second one is a reliable algorithm for shadow removing: it is based on the correlation between regions selected from the reference image and the current one. In addition, with our approach, the artifacts detected in presence of sudden light changes are removed. The experiments have been performed on real image sequences acquired both in indoor and outdoor environments with natural and artificial lights.