基于DNN的作业姿势评估方法及应用

doi:10.16265/j.cnki.issn1003-3033.2018.05.018

中国安全科学学报 ›› 2018, Vol. 28 ›› Issue (5): 105-110.doi: 10.16265/j.cnki.issn1003-3033.2018.05.018

基于DNN的作业姿势评估方法及应用

熊若鑫¹, 宋元斌¹ 高级工程师, 王宇轩²

1 上海交通大学船舶海洋与建筑工程学院,上海 200240;
2 东南大学交通学院,江苏南京 211189

收稿日期:2018-02-10 修回日期:2018-04-22 出版日期:2018-05-28 发布日期:2020-11-25
作者简介:熊若鑫(1994—),男,湖北黄冈人,硕士研究生,主要研究方向为机器视觉、行为安全。E-mail:xiongrxchn@sjtu.edu.cn。
基金资助:
国家自然科学基金资助(71271137);浦东新区卫计委重点学科群建设项目(PWZxq2017-16);上海交通大学文理交叉基金资助(15JCMY09)。

Deep neural network based posture assessment method and its application

XIONG Ruoxin¹, SONG Yuanbin¹, WANG Yuxuan²

1 School of Naval Architecture,Ocean & Civil Engineering,Shanghai Jiaotong University,Shanghai 200240,China;
2 School of Transportation,Southeast University,Nanjing Jiangsu 211189,China

Received:2018-02-10 Revised:2018-04-22 Online:2018-05-28 Published:2020-11-25

摘要/Abstract

摘要： 为有效预防工人肌肉骨骼疾病,提出基于深度神经网络(DNN)的作业姿势评估方法。利用人体关节点估计DNN模型,检测现场视频中工人的身体姿势;通过提取的骨骼关节点的空间位置,计算不同身体部位的姿势角度,并依据快速全身评估法(REBA)中的身体姿势角度与危害程度的关系,自动评估工人作业姿势的风险水平;以建筑工人典型施工作业姿势为例,验证该方法的有效性和实用性。结果表明:该方法能够克服身体遮挡、设备分辨率、光照条件等影响,准确检测关节点的位置;可自动连续评估工人的作业姿势,评估效率较高。

关键词: 职业健康, 肌肉骨骼疾病, 深度神经网络(DNN), 作业姿势, 快速全身评估法(REBA)

Abstract: In order to effectively prevent the occurrence of occupational musculoskeletal disorder,an approach was developed for assessing the risk levels of working postures based on deep neural network.The spatial locations of joints were estimated from on-site video through DNN,then the REBA was employed based on the calculating limb angels for the ergonomic analysis.The method was verified by analyzing typical construction working postures.The results show that the developed methodology can automatically and continually analyze sequence of postures and satisfies the recording conditions like dim light,occlusion of the partial body and low resolution of camera,and that the developed approach improves the traditional REBA via DNN,leading to higher efficiency of assessment for improving professional health of workers.

Key words: occupational health, musculoskeletal disorder, deep neural network (DNN), work posture, rapid entire body assessment

中图分类号:

X912.9

熊若鑫, 宋元斌, 王宇轩. 基于DNN的作业姿势评估方法及应用[J]. 中国安全科学学报, 2018, 28(5): 105-110.

XIONG Ruoxin, SONG Yuanbin, WANG Yuxuan. Deep neural network based posture assessment method and its application[J]. China Safety Science Journal, 2018, 28(5): 105-110.

参考文献

[1] National Occupational Research Agenda (NORA) Musculoskeletal Disorders Team.National occupational research agenda for musculoskeletal disorders:research topics for the next decade[R],2001.
[2] MARRAS W S,CUTLIP R G,BURT S E,et al.National occupational research agenda (NORA) future directions in occupational musculoskeletal disorder health research[J].Applied Ergonomics,2009,40(1):15-22.
[3] 邓乾旺,王泳朝,王长文.汽车装配工人肌肉骨骼疾患危害程度评价研究[J].中国安全科学学报,2014,24(4):50-55.
DENG Qianwang,WANG Yongchao,WANG Changwen.Study on harm degree evaluation of musculoskeletal disorders among worker in automobile assembly plan[J].China Safety Science Journal,2014,24(4):50-55.
[4] 何丽华,苏艳,曹磊,等.石油钻井行业工人肌肉骨骼疾患及影响因素分析[J].中华劳动卫生职业病杂志,2011,29(3):163-166.
HE Lihua,SU Yan,CAO Lei,et al.The analysis of factors influencing the work related musculoskeletal disorders of oil drillers[J].Chinese Journal of Industrial Hygiene and Occupational Diseases,2011,29(3):163-166.
[5] DAVID G C.Ergonomic methods for assessing exposure to risk factors for work-related musculoskeletal disorders[J].Occupational Medicine,2005,55(3):190-199.
[6] CHEN Jiayu,QIU Jun,AHN C.Construction worker's awkward posture recognition through supervised motion tensor decomposition[J].Automation in Construction,2017,77:67-81.
[7] CHENG Tao,VENUGOPAL M,TEIZER J,et al.Performance evaluation of ultra wideband technology for construction resource location tracking in harsh environments[J].Automation in Construction,2011,20(8):1173-1184.
[8] LIU Yanfei,CARLOS P R.Detection of body posture with low-cost CMOS camera systems for applications in medical care[C].International Conference on Electro/Information Technology,2009:301-306.
[9] PARK M W,BRILAKIS I.Construction worker detection in video frames for initializing vision trackers[J].Automation in Construction,2012,28:15-25.
[10] HIGNETT S,MCATAMNEY L.Rapid entire body assessment[J].Applied Ergonomics,2000,31(2):201-205.
[11] SIMONYAN K,ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[J/OL].Computing Research Repository,2014,(abs/1409.1556).http://dx.doi.org/10.1016/j.infsof.2008.09.005.
[12] CAO Zhe,SIMON T,WEI Shihen,et al.Realtime multi-person 2d pose estimation using part affinity fields[C].Computer Vision & Pattern Recognition,2017:1302-1310.
[13] ANDRILUKA M,PISHCHULIN L,GEHLER P,et al.2D human pose estimation:new benchmark and state of the art analysis[C].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2014:3 686-3 693.
[14] 王春雪,吕淑然.建筑施工作业安全感影响因素研究[J].中国安全科学学报,2016,26(5):13-18.
WANG Chunxue,LYU Shuran.Research on factors affecting sense of safety in construction operations[J].China Safety Science Journal,2016,26(5):13-18.

[1]	杨黎霞, 许茂增, 陈仁祥. 数据样本有限的交通恐怖袭击行为识别^*[J]. 中国安全科学学报, 2021, 31(8): 30-37.
[2]	佟瑞鹏, 李虹玮, 张娜, 许素睿, 安宇. 新工科视角下安全心理学研究进展与实践应用[J]. 中国安全科学学报, 2020, 30(1): 14-20.
[3]	徐胜, 金龙哲, 徐明伟. 受限空间钻孔工人的肌肉疲劳特性研究[J]. 中国安全科学学报, 2019, 29(3): 76-81.
[4]	吴小笛, 王海波, 张乐, 陈俞鹏, 薛朝军. 航空装配抗振外骨骼减振性能试验研究[J]. 中国安全科学学报, 2019, 29(11): 129-134.
[5]	师睿鑫. 基于图像处理的轮轴探伤系统对铁路安全的控制研究[J]. 中国安全科学学报, 2018, 28(S1): 22-28.
[6]	孙震. 高铁单司机职业健康分析[J]. 中国安全科学学报, 2018, 28(S1): 98-102.
[7]	李小冬，沈诚，王剑. 大型建筑施工企业职业健康管理状况调研[J]. 中国安全科学学报, 2013, 23(3): 155-.
[8]	游波，吴超，杨冕. 安全生理学原理及其体系研究[J]. 中国安全科学学报, 2013, 23(12): 9-.
[9]	陈全，温贺，陈波. 基于二元语义的职业健康安全管理体系绩效评价[J]. 中国安全科学学报, 2011, 21(10): 156-.
[10]	徐向东，赵宏展. 企业系统化安全管理的主要问题及成因分析[J]. 中国安全科学学报, 2010, 20(8): 142-.
[11]	张秋秋. 基于经济学视角的职业健康规制问题研究[J]. 中国安全科学学报, 2010, 20(5): 17-.
[12]	段淼，王起全，严琳. 职业健康安全管理体系运行中若干问题的探讨[J]. 中国安全科学学报, 2010, 20(3): 131-.
[13]	邹小伟，邓铁军，周祥蕴. 打造零事故的建筑安全文化[J]. 中国安全科学学报, 2009, 19(6): 77-.
[14]	刘小勇，王晓明. 特大工程建设项目引入职业健康安全管理体系的研究与应用[J]. 中国安全科学学报, 2009, 19(5): 56-.
[15]	李仲学，李翠平，刘双跃. 矿山安全法规标准与监管体系的国内外对比分析及其启示[J]. 中国安全科学学报, 2009, 19(3): 55-.

基于DNN的作业姿势评估方法及应用

Deep neural network based posture assessment method and its application

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价