Moving object detection with an adaptive background model

Background modeling is a critical case for background-subtraction-based approaches and also for a wide range of applications. A background generation becomes difficult when the scene is complex or an object stay for more than half of the time in the scene. In this paper, we propose a block-based scene background initialization and modeling with low computational cost which making them feasible for Embedded Platform. In general, many background subtraction approaches are sensitive to sudden illumination changes in the scene and does not update the background model properly over time. The proposed background modeling approach analyzes the illumination change problem. From the quantitative evaluation selected through a suite of metrics, and compared results obtained by some existing methods, our approach is effective for background generation.