Research on human action recognition and contrast based on OpenPose

YANG Jun; ZHANG Sujun; ZHANG Chuanghao; HUANG Jingjing

doi:10.13873/J.1000-9787(2021)01-0005-04

Chinese

您当前的位置：

首页 >

文章列表页 >

Research on human action recognition and contrast based on OpenPose

更新时间：2025-09-11

- Research on human action recognition and contrast based on OpenPose
- Vol. 40, Issue 1, Pages: 5-8(2021)
- 作者机构：
  
  武汉科技大学信息科学与工程学院
- 作者简介：
- 基金信息：
- DOI：10.13873/J.1000-9787(2021)01-0005-04
  CLC： TP391.41
- Published：2021
- 稿件说明：
移动端阅览
[1]杨君,张素君,张创豪,黄晶晶.基于OpenPose的人体动作识别对比研究[J].传感器与微系统,2021,40(01):5-8.

YANG Jun, ZHANG Sujun, ZHANG Chuanghao, et al. Research on human action recognition and contrast based on OpenPose[J]. 2021, 40(1): 5-8.
[1]杨君,张素君,张创豪,黄晶晶.基于OpenPose的人体动作识别对比研究[J].传感器与微系统,2021,40(01):5-8. DOI： 10.13873/J.1000-9787(2021)01-0005-04.

YANG Jun, ZHANG Sujun, ZHANG Chuanghao, et al. Research on human action recognition and contrast based on OpenPose[J]. 2021, 40(1): 5-8. DOI： 10.13873/J.1000-9787(2021)01-0005-04.

摘要

针对经典的人体动作对比识别方法有安装繁琐

设备需求量大

交互性不强

成本较高

数据量较大等不足

采用OpenPose可以提高动作识别对比的准确度

简易度以及交互性。利用OpenPose算法进行数据分析和挖掘

检测出图像中人体关键点信息并计算距离

再计算出夹角

判断人体动作的不足之处并提示反馈。实验结果表明:该系统安装简单

设备需求量减小、交互性增强、成本降低、数据量减小

对噪声干扰有较强的鲁棒性。

Abstract

Aiming at the shortcomings of classic human motion comparison and recognition method

the cumbersome installation

large demand for equipment

not strong interaction

high cost

and large amount of data

use OpenPose to improve the accuracy

simplicity and interactivity of the comparison of motion recognition.Use the OpenPose algorithm for data analysis and mining

detect the key point information of the human body in image and calculate the distance

then calculate the included angle

determine the inadequacies of the human body movement and prompt feedback.The experimental results show that the system requires simple equipment installation

reducs demand

enhances interactivity

reduces cost and data quantity

and it has strong robustness to noise interference.

关键词

Keywords

references

基于深度图像的实时多人体姿态估计 [J]. 肖贤鹏,刘理想,胡莉,张华. 传感器与微系统 . 2020(06)

基于运动传感器的手势识别 [J]. 王万良,杨经纬,蒋一波. 传感技术学报 . 2011(12)

基于三维加速度传感器的上肢动作识别系统 [J]. 王昌喜,杨先军,徐强,马祖长,孙怡宁. 传感技术学报 . 2010(06)

模板匹配技术在图像识别中的应用 [J]. 田娟,郑郁正. 传感器与微系统 . 2008(01)

基于BP神经网络的人脸识别技术的研究 [D]. 周昕. 南京理工大学 2007

基于粒子滤波算法的SAM模糊神经网络训练 [A]. 董国成,刘红云,周标准,崔艳,宋培.第三十一届中国控制会议论文集C卷[C]. 2012

Suitability of commercial barometric pressure sensors to distinguish sitting and standing activities for wearable monitoring [J] . F. Massé,A.K. Bourke,J. Chardonnens,A. Paraschiv-Ionescu,K. Aminian. Medical Engineering and Physics . 2014

Views

337

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Research progress of surface electromyography muscle strength estimation model

Occluded pedestrian detection method based on high-level semantic feature pyramid

Research on surgical instrument recognition method based on YOLOv5 and attention mechanism

Research on gas sensor humidity compensation based on GA-BP neural network

Head and shoulder posture estimation in sitting posture based on neural networks

Related Author

YU Fengfan

WEI Dejian

FENG Yanyan

MA Yifan

LI Zhenjiang

CHEN Tong

LI Qiming

GUO Feifei

Related Institution

College of Intelligence and Information Engineering,Shandong University of Traditional Chinese Medicine

Advanced Manufacturing College,Fuzhou University

Quanzhou Institute of Equipment Manufacturing,Haixi Institutes,Chinese Academy of Sciences

Quanzhou Vocational and Technical University

School of Mechanical and Automotive Engineering,Shanghai University of Engineering Science

⁰