Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Crowd behavior recognition method based on subgroup division and dynamic and static feature fusion

A feature fusion and recognition method technology, applied in cross-fields, can solve the problems of high noise sensitivity and poor effect, and achieve the effect of reducing complexity

Inactive Publication Date: 2020-06-12
NANJING UNIV OF POSTS & TELECOMM
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In practice, the momentum feature constructed for a single pedestrian target is not effective for the crowd gathering and is highly sensitive to noise.

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
  • Crowd behavior recognition method based on subgroup division and dynamic and static feature fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] Below in conjunction with accompanying drawing and specific embodiment, further illustrate the present invention, should be understood that these examples are only for illustrating the present invention and are not intended to limit the scope of the present invention, after having read the present invention, those skilled in the art will understand various aspects of the present invention All modifications of the valence form fall within the scope defined by the appended claims of the present application.

[0034] Such as figure 1 Shown is a crowd behavior recognition method based on subgroup division and dynamic and static feature fusion, including the following steps:

[0035] A video of a dense group with human behavior is collected, the user inputs the video, and the video is decomposed into video segments with 15 frames, wherein the interval of each video segment is 5 frames;

[0036] Take the first 20 frames of video, use each pedestrian as a feature point R, and...

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 crowd behavior recognition method based on sub-group division and dynamic and static feature fusion, and the method comprises the steps: firstly obtaining the spatial and temporal information of a moving target in a video image frame through employing an angular point tracking and background modeling method, and dividing spatially adjacent crowds into a plurality of sub-groups through employing the spatial region information of group distribution; secondly, on the basis of sub-group segmentation, obtaining sub-groups; three momentum characteristics of crowd movement are extracted; fusing the CNN neural network with a static sequence generated through the CNN neural network by utilizing Cholesky transformation; video classification is realized by using a GRU neuralnetwork, feature vectors are converted into crowd behavior tags by using an output function, and training video clips are marked as different description vocabularies by using a manual marking methodaccording to differences of behavior occurrence subjects, behavior occurrence places and behaviors.

Description

technical field [0001] The invention relates to a human behavior recognition method based on subgroup division and dynamic and static feature fusion, and belongs to the cross technical fields of behavior recognition, machine learning and the like. Background technique [0002] Group-level activity recognition has increasingly become a hot issue in the field of computer vision, and has a wide range of applications in intelligent video surveillance, public security, and sports competitions. In the feature extraction of crowd behavior analysis, most of the current feature extraction methods are based on a single pedestrian in the crowd as the target to establish momentum features, and then linearly combine the momentum features of each person into a feature vector for classification. In practice, the momentum feature constructed for a single pedestrian target does not work well for the situation of dense crowds, and is highly sensitive to noise. Contents of the invention [...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06N3/04G06N3/08
CPCG06N3/08G06V40/20G06V40/10G06V20/41G06N3/045
Inventor 葛宇轩陈志岳文静谢子凡王多崔明浩周传
Owner NANJING UNIV OF POSTS & TELECOMM
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
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