Human body action real-time identification method based on infrared image

A technology of infrared images and human actions, which is applied in character and pattern recognition, instruments, biological neural network models, etc., can solve the problems of low recognition rate and poor real-time performance of actions, so as to reduce the amount of calculation, improve the accuracy of extraction, and improve the accuracy of extraction. real-time effect

Active Publication Date: 2019-10-22
CHENGDU UNIVERSITY OF TECHNOLOGY
View PDF8 Cites 26 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a real-time human action recognition method based on infrared images to solve t

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
  • Human body action real-time identification method based on infrared image
  • Human body action real-time identification method based on infrared image
  • Human body action real-time identification method based on infrared image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0061] A preferred embodiment of the present invention provides a method for real-time recognition of human body movements based on infrared images, such as figure 1 shown, including the following steps:

[0062] Step 1: Construct the infrared image human body posture extraction network and the action recognition network based on bone features.

[0063] The network structure of infrared image human pose extraction is composed of the basic network MS-RsNet and the detection network of CenterNet framework. in:

[0064] (1) Multi-scale ResNet network-MS-ResNet

[0065] The multi-scale ResNet network provides input for subsequent action recognition by detecting and extracting the dynamic skeleton features of the human body in the filtered image (obtained by preprocessing the image). At present, among the frameworks for human body posture extraction, the representative ones are OpenPose, DensePose, and AlphaPose. The above-mentioned frameworks have achieved good detection result...

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 human body action real-time identification method based on an infrared image, and relates to the technical field of human body action recognition, and the method comprises the following steps: constructing an infrared image human body posture extraction network and an action recognition network based on skeleton features; acquiring an infrared pedestrian image data set and preprocessing the infrared pedestrian image data set, training an infrared image human body posture extraction network based on the preprocessed infrared pedestrian image data set, and obtaining a human body dynamic skeleton feature image; segmenting a region-of-interest sequence of the human body dynamic skeleton feature image, and training an action recognition network by utilizing a segmentation result; and acquiring a to-be-recognized infrared image, and performing action classification and prediction on the preprocessed to-be-recognized infrared image based on the infrared image human body posture extraction network and the action recognition network. According to the invention, the problems of poor real-time performance and low recognition rate of an existing behavior recognition method generally aiming at a visible light environment and performing human body behavior action recognition through an infrared image in a dark environment at night or in a severe weather environmentare solved.

Description

technical field [0001] The invention relates to the technical field of human motion recognition, in particular to a real-time human motion recognition method based on infrared images. Background technique [0002] The imaging principle of the infrared thermal imaging system is infrared spectral radiation imaging, which does not depend on the light source, is less affected by the weather, and has a long detection distance. It has strong application value in the fields of target recognition and detection, search and rescue, military, and driving assistance in the dark environment at night. . With the rapid development of machine vision and artificial intelligence, some breakthroughs have been made in the fields of image restoration, target tracking, target detection and recognition, and semantic segmentation in infrared thermal imaging images. However, there are still few studies on the intelligent recognition and analysis of human behavior and actions using infrared thermal ...

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): G06K9/00G06K9/62G06N3/04
CPCG06V40/20G06N3/045G06F18/241G06F18/214
Inventor 易诗谢家海
Owner CHENGDU UNIVERSITY OF TECHNOLOGY
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