Theme mining and behavior analysis method for video moving target

A moving target and behavior analysis technology, applied in the field of image processing, can solve the problem of inability to accurately describe the key summary information such as the target's moving behavior, and achieve the effect of accurate capture

Active Publication Date: 2021-02-09
CHINASO INFORMATION TECH
View PDF14 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although the invention discloses a video summary generation method based on target detection, this method only recognizes the target in the entire

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
  • Theme mining and behavior analysis method for video moving target
  • Theme mining and behavior analysis method for video moving target
  • Theme mining and behavior analysis method for video moving target

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0058]The first embodiment is a topic mining and behavior analysis method for video moving targets, such asfigure 1 As shown, it includes the following steps:

[0059]S1) Obtaining a video frame sequence and extracting the feature matrix Y of the video frame sequence includes the following steps:

[0060]S11) Obtain the video frame sequence I(x,y,t) containing the motion of the video moving target, and divide the video frame sequence I(x,y,t) into N video frame segments, where x and y respectively represent the spatial dimensions X coordinate and y coordinate, t represents time;

[0061]S12) Perform Gaussian convolution on the video frame sequence I(x,y,t) to obtain the Gaussian convolution result of the video frame sequence I(x,y,t)Where σ2,τ2Are the variances of the spatial and temporal dimensions of the video frame sequence I(x,y,t), f(·) is the mapping function that maps the video frame sequence I(x,y,t) to the corresponding pixel in the image sequence ;

[0062]S13) Calculate the three-dim...

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 relates to the field of image processing, and discloses a theme mining and behavior analysis method for a video moving target, and the method comprises the following steps: S1), obtaining a video frame sequence, and extracting a feature matrix of the video frame sequence; s2) performing topic mining by utilizing the feature matrix to obtain a topic matrix; and S3) performing behavioranalysis on the video frame sequence by using the topic matrix to obtain a behavior category of the video moving target. According to the invention, the image area with significant change in the video frame is extracted, the video expression is further constructed, and the motion attribute of the target in the video can be accurately captured. Besides, a streaming-based non-negative matrix factorization algorithm is adopted, a solving function is optimized by utilizing a weight matrix and a constraint condition, a theme abstract is mined, and time correlation information between video framesis expressed more accurately. The invention further provides a behavior multi-classification model based on the double-flow convolutional network, behavior labels for the mined topics are obtained, and the classification accuracy is improved.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a subject mining and behavior analysis method for video moving objects. Background technique [0002] In recent years, with the rapid development of the Internet, a large amount of video information has gradually become an important medium for people to perceive external things, such as live broadcast platforms, monitoring streams, etc. Due to the complexity of scenes and content targets in the video and the length of the video, and the video itself cannot provide an overview like text, and the video itself also contains a large amount of redundant blank information, it is difficult for people to efficiently analyze a video source in a short period of time. Analysis requires a lot of manpower and time. Generally, when the main target in the video is in motion, it can better reflect the information of the video itself. Therefore, how to accurately mine the target and motion attrib...

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): G06K9/00G06K9/46G06K9/62G06N3/04
CPCG06V20/42G06V10/464G06V10/50G06N3/045G06F18/253
Inventor 滕辉龙飞
Owner CHINASO INFORMATION TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap