Research And Implementation Of Distributed Intelligent Video Surveillance System

Video surveillance system has broad perspectives in application. The traditional video surveillance systems have a lot of deficiencies and are not suitable for setting up a large-scale network video surveillance system. This thesis focuses on distributed intelligent video surveillance system. After in-depth analysis and research of the architecture of the system, a feasible architecture of distributed intelligent video surveillance system is proposed. In order to solve the performance problem of video surveillance center and insufficient bandwidth issues, new video surveillance system uses embedded video surveillance front-end to process some surveillance function, such as intelligent video analysis. In this thesis, we design and implement a new distributed intelligent video surveillance system which includes embedded video surveillance front-end and video surveillance platform. The system provides remote video surveillance service for users in the way of C/S and B/S. At the same time, users can rely on the management platform to manage the system.In this paper, we propose an self adaptive transmission contorl approach based on QoE measurement of invalid data rate to make up the unreliability of data transmission in IP network. Also, protection of critical data transmission and keep-alive mechanism has been use in this system to overcome the shortcomings of existing data transfer protocol and enhance the system reliability.The experiment results show that video surveillance system is able to meet the demand of remote video surveillance service for users better. And the system in this thesis has got a great improvement of data transmission and system reliability. All of these make distributed intelligent video surveillance system have a high value of research and applies.