Mass surveillance system based on crowd sensing

OData support
Supervisor:
Dr. Simon Vilmos
Department of Networked Systems and Services

Nowadays mass events have huge number of visitors, however they carry countless danger as well. In large crowd a minor panic (prediction and prevention of them is a serious task) can cause unpredictable consequences, besides that the organizers do everything for participants’ safety. My solution is an integrated mass surveillance system, which shows a comprehensive picture of the state and movement of the crowd in near real time, thus helps the work of safety organizations. With new information they can react more quickly and precisely so lives can be saved either.

Most of the participants of today’s mass events already have a mobile device (smartphone, tablet) which contains different sensors such as GPS, gyroscope. We can collect data from these sensors, thus we also are able to obtain information from them. Because of this, it is possible to monitor the dynamics of mass, or to make estimates about the future states. This can be very useful in several ways. In case of a big event, although there are ideas and inaccurate estimates of distribution of mass, we can not exactly know how many people staying in one area at a given moment. If we could measure it, we would be able to better redeploy police units and dynamically modify evacuation plans if the need arises. The solution also allows us to send messages for participants depends on their position. These messages may include public notices, and even promotional messages. For the implementation I developed a mass surveillance model which in addition to being able to provide proper abstraction it can significantly reduce the amount of network communication and number of irrelevant data.

Downloads

Please sign in to download the files of this thesis.