Segmentation and tracking are two important aspects in visual surveillance systems. Manybarriers such as cluttered background, camera movements, and occlusion make the robustdetection and tracking a difficult problem, especially in case of multiple moving objects. Objectdetection in the presence of camera noise and with variable or unfavourable luminanceconditions is still an active area of research. This paper proposes a framework which caneffectively detect the moving objects and track them despite of occlusion and a priori knowledgeof objects in the scene. The segmentation step uses a robust threshold decision algorithm whichuses a multi-background model. The video object tracking is able to track multiple objects alongwith their trajectories based on Continuous Energy Minimization. In this work, an effectiveformulation of multi-target tracking as minimization of a continuous energy is combined withmulti-background registration. Apart from the recent approaches, it focus on making use of anenergy that corresponds to a more complete representation of the problem, rather than one thatis amenable to global optimization. Besides the image evidence, the energy function considersphysical constraints, such as target dynamics, mutual exclusion, and track persistence. Theproposed tracking framework is able to track multiple objects despite of occlusions underdynamic background conditions.
展开▼