|本期目录/Table of Contents|

[1]韩豫,李康,刘泽锋.基于场景理解的施工临边坠落险兆智能识别方法*[J].中国安全生产科学技术,2024,20(2):44-51.[doi:10.11731/j.issn.1673-193x.2024.02.006]
　HAN Yu,LI Kang,LIU Zefeng.Intelligent identification method of construction edge falling near-miss based on scene understanding[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2024,20(2):44-51.[doi:10.11731/j.issn.1673-193x.2024.02.006]
点击复制

基于场景理解的施工临边坠落险兆智能识别方法*
分享到：

《中国安全生产科学技术》[ISSN:1673-193X/CN:11-5335/TB]

卷:
20
期数:
2024年2期
页码:
44-51
栏目:
学术论著
出版日期:
2024-02-29

文章信息/Info

Title:
Intelligent identification method of construction edge falling near-miss based on scene understanding
文章编号:
1673-193X(2024)-02-0044-08
作者:
韩豫; 李康; 刘泽锋
（1.江苏大学 土木工程与力学学院，江苏 镇江 212013；
2.江苏大学 应急管理学院，江苏 镇江 212013）
Author(s):
HAN Yu;  LI Kang;  LIU Zefeng
(1.Faculty of Civil Engineering and Mechanics, Jiangsu University, Zhenjiang Jiangsu 212013, China;
2.School of Emergency Management, Jiangsu University, Zhenjiang Jiangsu 212013, China)
关键词:
临边坠落; 场景理解; 深度学习; 知识图谱; 险兆推理
Keywords:
edge falling;  scene understanding;  deep learning;  knowledge graph;  near-miss inference
分类号:
X947
DOI:
10.11731/j.issn.1673-193x.2024.02.006
文献标志码:
A
摘要:
为更及时、更有效地预防施工临边坠落事故的发生，并弥补现有智能预警方法在场景理解方面的不足，融合深度学习与语义推理，提出1种险兆识别方法。该方法通过neo4j构建险兆知识图谱，将引入轻量级视觉Transformer的YOLOx模型识别工人的险兆行为，设计描述空间关系的IoU计算方法并使用Cypher推理语言进行险兆推理。研究结果表明：施工临边坠落各要素识别的平均精度达91%以上，且IoU计算与险兆推理准确率均为100%，模型识别效果与险兆推理效果较好，该方法总体满足精度和速度的识别要求。研究结果可为实现施工临边坠落险兆行为的精准识别和预警提供参考。
Abstract:
In order to prevent the construction edge fall accidents more timely and effectively, and make up for the shortcomings of existing intelligent early-warning methods in scene understanding, a near-miss identification method was proposed by integrating deep learning and semantic inference.This method first constructed a near-miss knowledge graph through neo4j, and then introduced the YOLOx model with a lightweight visual Transformer to identify the near-miss behavior of workers.An IoU calculation method for describing the spatial relationship was designed, and the Cypher inference language was used to carry out the near-miss inference.The results show that the average accuracy of identifying various elements of construction edge falling is over 91%, and both the accuracy of IoU calculation and near-miss inference are 100%.The model identification effect and near-miss inference effect are good.The method generally meets the identification requirements of accuracy and speed.The research results can provide reference for achieving the accurate identification and early-warning of construction edge falling near-miss behavior.

参考文献/References:

［1］中华人民共和国住房和城乡建设部.住房和城乡建设部办公厅关于2020年房屋市政工程生产安全事故情况的通报［EB/OL］.(2021-05-21)［2023-09-10］.http://www.mohurd.gov.cn/gongkai/zhengce /zhengcefilelib/202210/20221026_768565.html.
［2］LE C Y,BENGIO Y,HINTON G.Deep learning［J］.Nature,2015,521(7553):436-444.
［3］谢定坤.模态融合的施工现场工人不安全行为识别方法研究［D］.武汉:华中科技大学,2020.
［4］祝铭悦,牛梓儒,万勇，等.基于图像的地铁车站施工人员不安全行为识别研究［J］.河北工业科技,2023,40(1):27-32. ZHU Mingyue,NIU Ziru,WAN Yong,et al.Research on identification of unsafe behaviors of construction personnel in subway station based on images［J］.Hebei Journal of Industrial Science and Technology,2023,40(1):27-32.
［5］程明皓.基于改进YOLO算法的加油站监控场景目标检测研究［D］.大庆:东北石油大学,2023.
［6］SON H,KIM C.Integrated worker detection and tracking for the safe operation of construction machinery［J］.Automationin Construction,2021,126:103670.
［7］程昱昊.基于轻量型神经网络的矿工不安全行为识别算法研究［D］.徐州：中国矿业大学,2022.
［8］VISHWAKARMA D K，KAPOOR R.Hybrid classifier based human activity recognition using the silhouette and cells［J］.Expert Systems with Applications,2015,42(20):6957-6965.
［9］MAITY S，BHATTACHARJEE D，CHAKRABARTI A.A novel approach for human action recognition from silhouette images［J］.IETE Journal of Research,2017,63(2):160-171.
［10］TANG Z，GU R，HWANG J N.Joint multi-view people tracking and pose estimation for 3D scene reconstruction［C］//2018 IEEE International Conference on Multimedia and Expo （ICME）,2018:1-6.
［11］陈敏.基于轨迹特征和深度学习的视频人体行为识别研究［D］.镇江：江苏大学,2023.
［12］AN L，TSOU M H，CROOK S E S，et al.Space-time analysis：concepts，quantitative methods，and future directions［J］.Annals of the Association of American Geographers,2015,105(5):891-914.
［13］GOUTHAMANKV,NAMBIAR A,SRINIVA K S,et al.Linguistically-aware attention for reducing the semantic gap in vision-language tasks［J］.Pattern Recognition,2021,112:107812.
［14］张旭华.框架语义推理技术研究［D］.太原:山西大学,2016.
［15］中华人民共和国住房和城乡建设部.建筑施工高处作业安全技术规范：JGJ 80—2016［S］.北京：中国建筑工业出版社，2016.
［16］中华人民共和国住房和城乡建设部.建筑施工易发事故防治安全标准：JGJ/T 429—2018［S］.北京：中国建筑工业出版社，2018.
［17］张笑非,黄智升,王东升,等.基于空间推理的城市路网交叉口模式研究［J］.道路交通与安全,2014，14（1）：31-35. ZHANG Xiaofei,HUANG Zhisheng,WANG Dongsheng,et al.Spatial reasoning based research of road network intersection patterns in cities［J］.Road Traffic and Safety,2014，14（1）：31-35.
［18］ZHIDCHENKO TV,SEREDINA M N,UDINTSOVA N M,et al.Design of energy-loaded systems using the Neo4j graph database［C］//IOP Conference Series:Earth and Environmental Science.IOP Publishing,2021,659(1):012108.

相似文献/References:

备注/Memo

备注/Memo:
收稿日期： 2023-09-25
* 基金项目： 国家自然科学基金项目（72071097）；教育部人文社会科学研究规划基金项目（20YJAZH034）；江苏大学应急管理学院专项科研项目（KY-B-10）；2023年江苏省研究生科研创新计划项目（KYCX23_3674）
作者简介： 韩豫，博士，教授，主要研究方向为智慧建造、行为安全与施工安全。
通信作者： 李康，硕士研究生，主要研究方向为智慧建造与施工安全。
更新日期/Last Update: 2024-03-11