Design And Implementation Of Intelligent Video Analysis Server Clusters

In recent years, with the development of artificial intelligence, network communication technology, video storage technology and video encoding and decoding technology, the development goal of monitoring has gradually become networked, intelligent and real-time monitoring. In the wisdom of public security, transportation, anti-terrorism and other application scenarios, Intelligent video surveillance need to meet the requirements of real-time. With the expansion of the intelligent monitoring system, the load of intelligent video analysis server is increasing rapidly. Thus, intelligent video analysis server cluster system, which can meet the needs of practical application, becomes one of the important requirements in this field.This paper mainly focuses on the research of the server cluster system of the actual combat video analysis server. Firstly, the domestic and foreign research status of intelligent monitoring system is analyzed. According to the actual requirements of the public security system, puts forward the implementation model, design method and system of intelligent video server cluster system. Secondly,the problem of how to manage multiple tasks and the information interaction between the task management server and multiple video analysis server node agent modules are discussed. The task management server is used to receive and process the intelligent analysis task, multiple video analysis server node composed of intelligent video analysis service pool. When the user needs for intelligent video analysis, task management server is from intelligent video analysis in a service pool to obtaining the corresponding resource analysis. Each of the intelligent video analysis contains a communication proxy service module and intelligent scheduling algorithm module server node, and calls the related intelligent analysis algorithm to carry on the intelligence analysis, at the same time, the result of the calculation is returned to the end user through the stream media forwarding server. Thirdly, the design and implementation of each module of the intelligent video analysis server cluster system are described. The design scheme of intelligent video server cluster load balancing, scalability and availability is given. A corresponding analysis server cluster load balancing algorithm is adopted, it solve the performance load balancing problem of video analysis server. At last, the implementation of each module of the intelligent server cluster is developed and tested. The software implement the video analysis server resource monitoring report, video analysis task startup, shutdown and hang up, parameter transmission of video analysis tasks, analysis results are returned(analysis results for text images and real-time coding). Through the actual system testing, the design basically achieved the expected design requirements.