Distributed real-time video monitoring processing system based on ZooKeeper

Abstract

The invention discloses a distributed real-time video monitoring processing system based on a ZooKeeper. The system comprises video acquisition equipment, a ZooKeeper server farm, a memory database server, a real-time video processing server farm RVPS, a streaming media domain name service and a client. In the invention, the ZooKeeper server farm is used to realize high availability of a cluster, resource distribution and calculation and load balancing. The client sends an operation instruction to the real-time video processing server farm through the memory database server and can modify the operation instruction at any time. The RVPS accesses a real-time video stream from video acquisition equipment. Through polling a processing instruction stored in the memory database server, a real-time video is processed. The RVPS sends an original coded video to the client through an RTSP streaming media protocol and feeds back a processing result to the client through the memory database server so as to realize separation of the video and the processing result. The client accesses an RTSP video stream with an unchanged domain name and a changed IP address through the streaming media domain name service.

Description

technical field [0001] The invention belongs to the field of distributed real-time video processing, in particular to a distributed real-time video monitoring and processing system based on ZooKeeper. Background technique [0002] With the wide application of high-definition video surveillance equipment and technology in the field of public security, a large amount of video data is added every moment. Massive high-definition video has played a great role in speeding up the cracking of vicious criminal cases and maintaining social public security. However, at present, the analysis and processing of real-time video mainly rely on manpower, and more video data are stored on disk for later reference. In order to process massive real-time video more effectively, it is necessary to implement a simple, efficient, and easy-to-deploy distributed real-time video processing system. [0003] Existing parallel processing platforms or frameworks, such as Hadoop, are usually used for offl...

AI Technical Summary

Problems solved by technology

[0003] Existing parallel processing platforms or frameworks, such as Hadoop, are usually used for offline processing and analysis of historical data; real-time stream processing frameworks such as Storm and Spark-Streaming are suitable for discrete data processing, not continuous video streams

At the same time, the above-mentioned platforms or frameworks are bulky, difficult to deploy, and consume a lot of resources

Images