Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Monitoring video joint analysis method and system

A technology for monitoring video and joint analysis, applied in the direction of resource allocation, program control design, multi-programming device, etc., can solve the problems of slow analysis speed and low computing efficiency, to ensure computing speed, ensure analysis efficiency, and reduce computing consumption. Effect

Pending Publication Date: 2020-11-13
HUAZHONG UNIV OF SCI & TECH
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Aiming at the defects and improvement needs of the prior art, the present invention provides a monitoring video joint analysis method and system, the purpose of which is to solve the problems of slow analysis speed and low calculation efficiency existing in the existing monitoring video analysis technology

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Monitoring video joint analysis method and system
  • Monitoring video joint analysis method and system
  • Monitoring video joint analysis method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0082] A monitoring video joint analysis method, such as figure 1 shown, including:

[0083] In the current analysis period, the surveillance video stream that is predicted to contain the target scene is regarded as a potential video stream, and the surveillance video stream that is predicted to not contain the target scene is regarded as a non-potential video stream, respectively at the first sampling rate and the second sampling rate Sampling of latent and non-latent video streams;

[0084] Input the sampling results of the latent video stream and the non-latent video stream into the full-featured target detection model and the lightweight target detection model for target detection, and output the detection results of the full-featured target detection model in real time;

[0085] At the end of the current analysis period, analyze whether each video stream contains the target scene in the current analysis period according to the detection results of the full-featured targe...

Embodiment 2

[0138] A surveillance video joint analysis system, such as figure 1 shown, including: GPU, Receiver, Selector, and Analyzer;

[0139] The receiver is used to take the surveillance video stream predicted to contain the target scene as a potential video stream, and the surveillance video stream predicted not to contain the target scene as a non-potential video stream in the current analysis period, respectively with the first sampling Sampling the potential video stream and the non-potential video stream at a rate and a second sampling rate;

[0140] The selector is used to input the sampling results of the potential video stream and the non-potential video stream into the full-featured target detection model and the lightweight target detection model for target detection, and output the detection results of the full-featured target detection model in real time;

[0141] An analyzer, configured to analyze whether each video stream contains a target scene in the current analysis...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a monitoring video joint analysis method and system, and relates to the fields of big data system and analysis. The method comprises the steps: respectively sampling a potential video stream and a non-potential video stream at a higher sampling rate and a lower sampling rate in a current analysis period; respectively inputting sampling results of the potential video streamand the non-potential video stream into a full-feature target detection model and a lightweight target detection model for target detection, and outputting a detection result of the full-feature target detection model in real time; at the end of the current analysis period, analyzing whether each video stream contains a target scene in the current analysis period according to the detection resultsof the two target detection models, and predicting whether each video stream contains the target scene in the next analysis period in combination with the historical analysis result; the two target detection models run in the GPU, and the complexity and detection precision of the full-feature target detection model are high. According to the invention, the analysis speed and the calculation efficiency of the monitoring video can be effectively improved on the basis of ensuring the analysis precision.

Description

technical field [0001] The invention belongs to the field of big data systems and analysis, and more specifically relates to a monitoring video joint analysis method and system. Background technique [0002] Currently, more and more inexpensive surveillance cameras are being deployed in public places to monitor an area, manage traffic, detect unusual accidents and record clues. Quickly and accurately locating key scenarios (abnormal events or events that users care about) is the main goal of analyzing surveillance video. In the early days, these recorded massive video data were manually analyzed, which was not only error-prone, but also expensive. Target detection technology is an important part of computer vision. It determines whether there is a target in the field of view through image processing technology, and locates the position of the target. Because of its excellent analysis performance, target detection technology has been gradually applied to surveillance video a...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F9/50G06K9/00G06K9/34
CPCG06F9/5027G06F9/5022G06F9/5016G06V20/46G06V10/267
Inventor 曹强张宸
Owner HUAZHONG UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com