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

Crowd abnormal event detection method based on hybrid tracking and generalized linear model

A generalized linear model and abnormal event technology, applied in the field of image feature description, can solve problems such as poor detection effect and poor robustness

Active Publication Date: 2018-07-13
BEIJING UNION UNIVERSITY
View PDF9 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method has poor detection effect and poor robustness when the video quality is low and the crowd density is relatively dense.

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 abnormal event detection method based on hybrid tracking and generalized linear model
  • Crowd abnormal event detection method based on hybrid tracking and generalized linear model
  • Crowd abnormal event detection method based on hybrid tracking and generalized linear model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0068] Such as figure 1 As shown, step 100 is executed to extract pedestrian trajectories. As an important feature for detecting the moving direction and behavior of foreground objects, the motion trajectory of pedestrians plays an important role. In terms of pedestrian tracking, this step uses a hybrid tracking model based on zero termination constraints. In this model, the nearest neighbor search method is used to estimate the head position of the object and the HSIM algorithm is used to realize the vision-based tracking. In step 100, execute sub-step 101 to read monitoring images and reference points. First input a crowded video V and extract each frame in the video. Then V∈{v i ; 1 ≤ i ≤ n} as input to track human objects using a hybrid tracking model. Suppose the input video contains n frames, starting from the first frame v 1 Randomly find the reference point R in j to get the head part of the human body in the first frame. R j =(x j ,y j ); 1≤x j ≤M; 1≤y j ...

Embodiment 2

[0074] Mass incidents refer to incidents that are participated in or triggered by groups. Such incidents often take the form of gathering forces, have a great influence on society, have a strong impact, and have a major impact on the stable development of society. Such as image 3 , 3A , 3B, people walk freely in the video scene, and these normal life events are defined as group normal events, such as Figure 4 , 4A As shown in , 4B, crowds gather for a short time or flee in panic. In reality, these events are often accompanied by fights, riots, traffic accidents and natural disasters, which have adverse effects on society. We define it as a group unusual event.

[0075]Aiming at the problem that the current crowd abnormal event detection model has high computational complexity, a crowd abnormal event detection algorithm based on hybrid tracking and generalized linear model (GLM) is proposed. Define the input crowded video V to represent a sequence of frames, where V ∈ {v ...

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 provides a crowd abnormal event detection method based on a hybrid tracking and generalized linear model. The crowd abnormal event detection method based on the hybrid tracking and generalized linear model includes the following steps of tracking pedestrian trajectories, extracting feature points of tracking paths, and conducting classification of at least one direction and activityof foreground objects using neural networks based on the generalized linear model. The crowd abnormal event detection method based on the hybrid tracking and generalized linear model conducts modelling using the neural networks with the hybrid tracking model and a genetic algorithm combined, and then an established feature model is used for representing group event information in the scene more intuitively. Furthermore, selecting a appropriate feature model has an important influence on the performance of final group abnormal event detection.

Description

technical field [0001] The invention relates to the technical field of image feature description, in particular to a crowd abnormal event detection method based on hybrid tracking and generalized linear model. Background technique [0002] In recent years, the possibility of various major mass abnormal events has increased significantly, and public safety issues have received increasing attention from the state. As a key technology in the field of public security, intelligent video surveillance (intelligent video surveillance, IVS) has been extensively researched and popularized. This technology integrates many academic fields such as image processing, pattern recognition and artificial intelligence, and is one of the research hotspots in the computer field. [0003] Mass incidents refer to incidents that are participated in or triggered by groups. Such incidents often take the form of gathering forces, have a great influence on society, have a strong impact, and have a maj...

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): G06K9/00
CPCG06V20/42G06V20/53
Inventor 玄祖兴郭燕飞王海孙欣
Owner BEIJING UNION UNIVERSITY
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