Occlusion Detection and Handling In Video Surveillance

This paper highlights to fill the missing parts from the past history of frames that are being recorded if available, when occlusion is detected. Occlusion is the hiding of an object by another during multiple human tracking. It introduces a multiple human objects tracking system, which detects and tracks multiple objects in crowded scene with occlusion. For multiple object tracking, it is important to maintain the history of objects before and after occlusions. When tracking multiple objects, we separate the object state into three parts: Before, during and after occlusion. An Improved Mean Shift Tracking algorithm (IMST) is specially used for flocking occlusion targets. Occlusion can be detected by calculating the centre of mass of both the objects and when the distance between them is zero. By comparing the frames, the occluded part is identified and the missing part is filled from the matched frame when occlusion is detected.