A novel background modeling approach for accurate and real-time motion segmentation
Tang Wang, Chen Gong, He-Qin Zhou · 2006
Fast and accurate segmentation of moving objects in video sequences is a basic task in many computer vision and video analysis applications. This paper presents a novel method for background modeling based on the unscented Kalman filter (UKF) algorithm. In addition, shadow detection is conducted during motion segmentation. Experiments are performed under different conditions indoor and outdoor. The results show that the proposed algorithm is effective and efficient in background modeling and motion segmentation