Spatiotemporal subspace based video analysis for changes and stationarities in visual surveillance

Visual surveillance has been a very active research topic in the last few years due to its growing importance in security and law enforcement. More and more surveillance cameras are installed and the massive amount of data involved makes it infeasible to guarantee vigilant monitoring by human operators for long periods of time. As a result, video feeds are usually archived for forensic purposes in the event suspicious activities take place. In order to assist human operators with identification of important events in videos an “intelligent” visual surveillance system can be used. Such a system requires fast and robust methods for background initialization and moving object detection. Unfortunately, the existing motion detection approaches cannot handle various challenges associated with background initialization such as occlusion, camera jitter, and lighting changes. Furthermore, these conventional approaches fail to accurately detect certain types of motion patterns such as periodic and repeated motions.