Video structuring method, related equipment, system and storage medium

Abstract

The invention provides a video structuring method. The video structuring method comprises the following steps that a streaming media platform imports equipment information in advance; the streaming media platform receives the video instruction and generates a video stream; the application server receives the video stream; the application server generates a structuralization request and sends the structuralization request to the neural network server, so that the neural network server performs structuralization analysis on the video stream; and the application server displays and updates the structured progress in real time on the page, and can quickly retrieve the monitoring video of the current task after the structured progress is finished. The invention provides streaming media platformequipment, application server equipment, a video structuring system and a computer readable storage medium. Compared with the prior art, by adopting the technical scheme of the invention, the function of online structured analysis of the monitoring video into the searchable image and text data is realized, and the efficiency of looking up the monitoring video is high.

Description

technical field [0001] The present invention relates to the technical field of video image processing, in particular to a video structuring method, related equipment, system and storage medium. Background technique [0002] As network applications become more and more extensive, especially monitoring equipment applications are also increasing. Among them, the intelligence and structure of traditional monitoring equipment are becoming more and more important. Currently, surveillance on the market can be divided into two types in terms of video structure: one is an intelligent real-time structured camera, and the other is a non-intelligent ordinary camera. [0003] The solution adopted by the traditional non-intelligent monitoring real-time video structured server of the related technology is to collect and structure the real-time video stream through the edge computing engine and the neural network server. [0004] However, traditional non-smart cameras only have video moni...

AI Technical Summary

Problems solved by technology

[0004] However, traditional non-smart cameras only have video monitoring and recording functions, and lack the ability to perform structural analysis on the objects appearing in the video screen

And the current new generation of intelligent monitoring equipment is expensive, and the penetration rate is far lower than traditional monitoring

Smart cameras can structure the monitored video streams in real time, but smart cameras are products that have only appeared in recent years and have not yet been widely popularized. Most of the installed cameras in the society are non-intelligent ordinary cameras

Since the price of smart cameras is more than ten times or even dozens of times that of ordinary cameras, the replacement cost is high

If non-intelligent monitoring wants to be structured, it needs to manually download the video and upload it to the server for structured analysis. The operation is time-consuming

[0005] In view of the current problem that the existing non-intelligent monitoring cannot carry out structured video streaming, combined with the current practical problems: the number of non-intelligent monitoring is large, the distribution is wide, the replacement is difficult and expensive, and the incident only needs a certain time when the incident occurs Analyze the surveillance video of the segment and part of the surveillance area

A large number of existing traditional monitoring cannot perform effective target retrieval through structured analysis, and the cost of updating equipment or deploying real-time structured servers is high

Images