Background initialization for the task of video-surveillance

摘要

In this work, we propose a region-wise and batch processing approach for background initializationin video-surveillance based on a spatio-temporal analysis.First, the related work has been explored. Then, the efforts are focused on developing anew background initialization approach to outperform the literature performance. To this end,a temporal analysis and a spatial analysis are performed. In the first stage, we use a previouswork techniques adding motion information to increase performance. In the second stage, amultipath iterative reconstruction scheme is performed to build the true background under theassumption of background smoothness, i.e. the empty scene is smoother than the scene withforeground regions.Finally, the results over challenging video-surveillance sequences show the quality of theproposed approach against related work.