Research And Implementation Of Intelligent Video Surveillance System Based On RTSP

Video surveillance system is an important part of security system,with information and intuitive content, are widely used in various fields of agricultura, industrial, military and social life. With the development of network technology, embedded technology and streaming media technology, it improves gradually toward the direction of front-end integration, digitization, networking and intelligence.Modern video surveillance system has the characteristics of high-definition screen and real-time monitoring, requiring higher network bandwidth, therefore more efficient video coding technology is used to reduce transfering network data and streaming media technology is applied to play smoothly network video. Web technology design client to break the traditional monitoring centralized management mode and geographical restrictions in favour of expansion and upgrading. In order to improve efficiency of system and avoid drawbacks of human monitoring, computer vision technology is applied to inspect moving objects and track automaticly. Take security and stability, into consideration, php layer of security is added to isolation server core and web client, reducing the possibility of malicious attacks; it make user login authentication and user actions recording independent from server core.On account of high-definition screen and data transmission of high network bandwidth, network video streaming smooth playback, broader scope of monitoring, intelligent monitoring and system security and stability requirements in current video surveillance system; this paper proposes an intelligent video monitoring system based on RTSP streaming protocol with automatic tracking function and multiple alarm mechanism. Improved TFRC-FIT algorithm based on bit rate smoothing for adaptive QoS control module is proposed to ensure that the needs of video playback and network media stream jitter, and threshold window and multi-reference frame smooth steering gear control strategy is proposed to reduce servos jitter and improve the efficiency of steering gear control in the steering gear PTZ control section, a web server is deployed in the system using LAMP architecture; video data collection, transmission and H.264hardcoded are finished; system builds RTSP streaming media server for real-time monitoring and playback of queries; Web client embedded vlc Web plugin is designed using Web technology; frame difference method with the help of opencv computer vision library is used to detect object trajectories and then the algorithm is applied to application system to drive PTZ servo tracking a target moving objects; system control command transmission communication protocol is designed and php security layer is used to improve server security and complete user login authentication and user operation records; different abnormal cases trigger gprs remote alarm and buzzer alarm function; addition, cross-platform’s Web client is also research content of the follow-up.