[1]郭园,郭晨旭,时新,等.基于OpenPose学习坐姿分析的桌椅人机适应性研究[J].林业工程学报,2020,5(02):179-185.[doi:10.13360/ j.issn.2096-1359.201909016]
 GUO Yuan,GUO Chenxu,SHI Xin,et al.Adaptability of study desks and chairs based on analysis of sitting posture using OpenPose[J].Journal of Forestry Engineering,2020,5(02):179-185.[doi:10.13360/ j.issn.2096-1359.201909016]
点击复制

基于OpenPose学习坐姿分析的桌椅人机适应性研究()
分享到:

《林业工程学报》[ISSN:1001-8081/CN:32-1160/S]

卷:
5
期数:
2020年02期
页码:
179-185
栏目:
家具制造工程
出版日期:
2020-03-11

文章信息/Info

Title:
Adaptability of study desks and chairs based on analysis of sitting posture using OpenPose
文章编号:
2096-1359(2020)02-0179-07
作者:
郭园12郭晨旭2时新2申黎明1*
1. 南京林业大学家居与工业设计学院,南京 210037; 2. 重庆交通大学,重庆 400074
Author(s):
GUO Yuan12 GUO Chenxu2 SHI Xin2 SHEN Liming1*
1. College of Furnishings and Industrial Design, Nanjing Forestry University, Nanjing 210037, China; 2. Chongqing Jiaotong University, Chongqing 400074, China
关键词:
OpenPose 坐姿 学习桌椅 人机关系 适应性
Keywords:
OpenPose sitting posture study desk and chair ergonomics relation adaptability
分类号:
TS664
DOI:
10.13360/ j.issn.2096-1359.201909016
文献标志码:
A
摘要:
运用非接触式Kinect运动捕捉系统及OpenPose二维姿势检测开源实时系统,采集小学生坐姿行为中身体关键指标变化数据,通过分析各个关键指标在不同任务及时间段内的具体变化情况,总结人机适应的内在变化规律,以便为后期的桌椅适应性调节方案提供可靠依据。结果表明:颈部弯曲和视距受任务影响显著。书写任务下颈部弯曲均值大、视距小,表现出较强的振幅频率,且随时间延长有增长趋势。阅读任务和书写任务中的颈部弯曲超过20°的时间占比以及近距离用眼情况比较近似,而在平板电脑任务中,颈部弯曲数值则小于其他两项任务中的该数值,视距也相对更大,但躯干弯曲超过20°的时间增加,其振幅频率变得突出,说明躯干是使用平板电脑任务中变化性最大的身体部位。建议学习桌采用恰当方式引导小学生采用45°~60°竖立书本进行阅读,以减少颈部弯曲,加大视距。此外,可以适时调整桌高满足书写到使用平板电脑任务的变换,减缓躯干弯曲和振幅变化。
Abstract:
Using the non-contact Kinect motion capture system and two-dimensional posture detection open source real-time system(OpenPose), this study collected the data of key body indicators in sitting postures of primary school students, analyzed the specific changes of key indicators in different tasks and periods, and then summarized the inherent regulations of ergonomics adaptation, so as to provide reliable bases for the later adaptive adjustment of desks and chairs. The results showed that the neck flexion and visual distance were significantly affected by the tasks of writing, reading and using tablet PC(P < 0.05). In the writing task, the mean value of neck flexion became larger and the visual distance was shorter than the other two tasks, and the amplitude probability distribution function(APDF)was significant, which indicated that the neck was the most changeable body part in the writing task. In addition, the neck flexion tended to be increased as time went on in the writing and tablet PC tasks. In the reading and writing tasks, the time percentage of neck flexion exceeded 20° and visual distance were approximate, while in tablet PC task, the value of neck flexion was smaller than the former tasks, and the visual distance was relatively larger. However, the trunk flexion exceeded 20° and with the time increased, the amplitude probability distribution function(APDF)was significant, which indicated that the trunk was the most changeable body part in the tablet PC task. The mean value of trunk-thigh angle was the largest in reading task, which indicated that the body was relatively relaxed, but the value of neck flexion and visual distance was similar to the task of writing. It was suggested that the study desks should adapt to students to read by erecting books with the angles of 45°-60°, so that it would reduce neck flexion and lengthen visual distance. In addition, with the help of intelligent detection technology, according to the sitting posture data real-time acquired, the desk system adjusted height of desk in due course to meet the requirements from the writing task to the tablet PC task, which reduced trunk flexion and amplitude variation. For the unhealthy sitting posture, the desk system gave an audio reminder or other ways to help primary school students to maintain a healthy posture via the adaption of a dynamic balance.

参考文献/References:

[1] MANGHISI V M, UVA A E, FIORENTINO M, et al. Real time RULA assessment using Kinect v2 sensor[J]. Applied Ergonomics, 2017, 65: 481-491. DOI:10.1016/j.apergo.2017.02.015.
[2] 孙辛欣, 周頔, 姜斌, 等. 办公座椅坐姿行为的聚类与分析[J]. 林业工程学报, 2018, 3(5): 158-164. DOI:10.13360/j.issn.2096-1359.2018.05.024.
SUN X X, ZHOU D, JIANG B, et al. Clustering analysis of sitting posture on office chair[J]. Journal of Forestry Engineering, 2018, 3(5): 158-164.
[3] CAO Z, SIMON T, WEI S E, et al. Realtime multi-person 2D pose estimation using part affinity fields[C]∥2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR), July 21-26, 2017. Honolulu, HI. New York, USA: IEEE, 2017. DOI:10.1109/cvpr.2017.143.
[4] MAZHAR O, NAVARRO B, RAMDANI S, et al. A real-time human-robot interaction framework with robust background invariant hand gesture detection[J]. Robotics and Computer-Integrated Manufacturing, 2019, 60: 34-48. DOI:10.1016/j.rcim.2019.05.008.
[5] 唐心宇, 宋爱国. 人体姿态估计及在康复训练情景交互中的应用[J]. 仪器仪表学报, 2018, 39(11): 195-203. DOI:10.19650/j.cnki.cjsi.J1803879.
TANG X Y, SONG A G. Human pose estimation and its implementation in scenario interaction system of rehabilitation training[J]. Chinese Journal of Scientific Instrument, 2018, 39(11): 195-203.
[6] FRANSSON-HALL C, GLORIA R, KILBOM Å, et al. A portable ergonomic observation method(PEO)for computerized on-line recording of postures and manual handling[J]. Applied Ergonomics, 1995, 26(2): 93-100. DOI:10.1016/0003-6870(95)00003-u.
[7] FETTWEIS T, ONKELINX M N, SCHWARTZ C, et al. Relevance of adding a triangular dynamic cushion on a traditional chair: a 3D-analysis of seated schoolchildren[J]. Clinical Biomechanics, 2017, 49: 113-118. DOI:10.1016/j.clinbiomech.2017.09.002.
[8] MURPHY S, BUCKLE P, STUBBS D. Classroom posture and self-reported back and neck pain in schoolchildren[J]. Applied Ergonomics, 2004, 35(2): 113-120. DOI:10.1016/j.apergo.2004.01.001.
[9] 郑泽铭. 人的坐姿检测方法及行为劝导研究[D]. 杭州: 浙江大学, 2013.
ZHENG Z M. Sitting posture sensing and healthy behavior persuation[D]. Hangzhou: Zhejiang University, 2013.
[10] 王丽君, 李黎, 张帆. 基于3D运动捕捉系统的坐姿角度和舒适度研究[J]. 中南林业科技大学学报, 2013, 33(12): 146-150. DOI:10.14067/j.cnki.1673-923x.2013.12.009.
WANG L J, LI L, ZHANG F. Study of sitting posture angle and comfort based on 3D motion capture system[J]. Journal of Central South University of Forestry & Technology, 2013, 33(12): 146-150.
[11] GONÇALVES M A, AREZES P M. Postural assessment of school children: an input for the design of furniture[J]. Work, 2012, 41: 876-880. DOI:10.3233/wor-2012-0257-876.
[12] STRAKER L M, COLEMAN J, SKOSS R, et al. A comparison of posture and muscle activity during tablet computer, desktop computer and paper use by young children[J]. Ergonomics, 2008, 51(4): 540-555. DOI:10.1080/00140130701711000.
[13] GELDHOF E, DE CLERCQ D, DE BOURDEAUDHUIJ I, et al. Classroom postures of 8-12 year old children[J]. Ergonomics, 2007, 50(10): 1571-1581. DOI:10.1080/00140130701587251.
[14] CICCARELLI M, STRAKER L, MATHIASSEN S E, et al. ITKids Part Ⅱ: variation of postures and muscle activity in children using different information and communication technologies[J]. Work, 2011, 38(4): 413-427. DOI:10.3233/wor-2011-1168.
[15] 黄馨慧, 王明进, 何鲜桂, 等. 184名8~12岁儿童不同阅读距离及阅读时间调节滞后和调节反应的观察[J]. 川北医学院学报, 2017, 32(1): 33-37. DOI:10.3969/j.issn.1005-3697.2017.01.010.
HUANG X H, WANG M J, HE X G,et al. Observation of accommodative response and accommodative lag under different reading distance and reading time in 184 children aged 8-12 years[J]. Journal of North Sichuan Medical College, 2017, 32(1): 33-37.

备注/Memo

备注/Memo:
收稿日期:2019-09-14 修回日期:2019-10-14
基金项目:重庆市教委科学技术研究项目(KJ1705137)。
作者简介:郭园,女,副教授,研究方向为家具设计与工程。通信作者:申黎明,男,教授。E-mail:shenlimingda@hotmail.com
更新日期/Last Update: 2020-03-10