Classifier training method for behavior recognition

A training method and classifier technology, which is applied in the field of classifier training for behavior recognition, can solve problems such as many holes, large gaps, and inability to identify, and achieve the effects of good robustness, accurate description, and insensitive initial parameters

Inactive Publication Date: 2015-01-21
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF2 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, because the gap between two adjacent points in the sparse flow is too large and there are too many holes, it is impossible to effectively capture the spatio-temporal characteristics of the action in the sparse motion flow, especially in fast motion, pairwise STIP The gap between them will become very large, and it is basically impossible to describe the spatio-temporal characteristics of the action, and thus cannot be recognized

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

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0016] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the embodiments.

[0017] In this patent, a method based on transforming sparse local features into dense motion flows is proposed. Specifically, firstly, the Dollar detector is used to extract the spatio-temporal interest point STIP from the video, and the extracted STIP is used to construct a sparse motion flow. The Dollar detection is an action analysis detection operator proposed by Dollar et al. Gabor filter is composed of sine and cosine functions. The main function of the detection operator is to extract the position of local motion information from the motion video. Because the gap between two adjacent STIPs in the sparse motion flow is too large, the motion information of the action cannot be effectively captured through the sparse motion flow, especially in fast motion, the gap be...

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 classifier training method for behavior recognition, and belongs to the technical field of image processing. The method includes the steps that STIPs of an input movement video image stream are extracted according to a Dollar detection operator; partial cavities between every two STIPs are filled, and all pixel points of which the vertical distance to the straight lines formed by every two STIPs is smaller than a preset threshold are set as new STIPs; all the current STIPs are presented based on an LDPD descriptor, a statistical histogram of the current movement video image stream is formed based on the LDPD description vector quantity of each STIP, the histogram serves as a training sample, and a behavior recognition classifier is output based on a support vector machine. The training method is used for behavior recognition, is not sensitive to an initial parameter, and is good in robustness when used for behavior recognition.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a classifier training method for behavior recognition. Background technique [0002] Human action recognition in videos has become a highly concerned research area, and action recognition has been applied in various fields, including: video indexing and browsing, video surveillance, gesture recognition, sports event analysis, etc. Behavior recognition is mainly divided into behavior analysis and recognition. Only with good behavior analysis can better recognition be carried out. At present, although various research institutions continue to conduct research on human motion analysis, there are still many unsolved problems. This is because in the real world, similar actions can be made by objects of different sizes, appearances, speeds, and poses. In addition, occlusion of static or moving objects, lighting changes, or shadows can have a relatively large nega...

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): G06K9/66
CPCG06V40/23G06V20/40G06V30/194G06V2201/07
Inventor 解梅许茂鹏张碧武卜英家
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA
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