A fuzzy inference framework for detecting intrusions in urban transit

It is very important to prevent crimes, accidents, and incidents, so many surveillance systems are equipped in urban transit system. But in most current surveillance systems, supervisors have to monitor many screens continuously. Therefore, intelligent systems are needed by which those tedious monitoring tasks are done. These intelligent surveillance systems have two parts: image processing, context inference module. Because there are many uncertain events in urban transit, fuzzy inference engine is needed that efficiently handle these events and solve the problems that can occur in the dangerous situation. In this paper, we present a fuzzy framework that can efficiently detect dangerous situations in urban transit and classify the contexts according to their dangerous situation.