The Software Design Of Transaction Center Platform For Metropolitan Aera Video Surveillance System

With the fast development of economy in China, the process of urbanization is developing rapidly. Meanwhile, since social contradictions continue to arise and have a strong tendency to sharpen, the security problems during urbanization process are becoming more and more prominent. Video surveillance technology, which is capable to help urban managers to prevent and detect illegal and criminal activities efficiently, has become one of the most important methods to guarantee the safety of people life and property. Therefore, it is of significant importance to the sound development of cities and the construction of a harmonious society, by establishing a high-definition video monitoring system across the whole city.Based on the complex transaction management requirement of metro video surveillance systems, the transaction center platform software for the metropolitan area video surveillance system which runs on Linux platform has been researched and developed. This software has realized functions such as resource management, user management, privilege management, organization management, alarm management, recording and playback management, and service management. To meet the demand of multiple user privilege management, an authority management scheme based on organizations and roles is developed by realizing the permission assignment of users under different organizations via configuring different roles under each organization. Meanwhile, to settle the problems of long request response time and poor load balance degree, when scheduling video using conventional streaming media load balancing algorithm, a novel load balancing algorithm which combines the new minim-load-priority load balancing algorithm based on the similarity of video requests g and dynamic load migration is designed. Via the function and performance tests, the software developed is confirmed to have met the system requirements. What’s more, the novel load balancing algorithm designed in this paper can shorten the response time of requests and reduce the cluster load variance, in comparison with the traditional light-load-priority algorithm at typical playback scales of metro video surveillance systems. Thus results in an improved load balancing degree and an enhanced system resource usage.