|本期目录/Table of Contents|

[1]唐豪,奉鑫鑫,高曙,等.基于视频分析的空管员违规行为识别方法*[J].中国安全生产科学技术,2023,19(1):196-201.[doi:10.11731/j.issn.1673-193x.2023.01.029]
　TANG Hao,FENG Xinxin,GAO Shu,et al.Violations recognition method of air traffic controllers based on video analysis[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2023,19(1):196-201.[doi:10.11731/j.issn.1673-193x.2023.01.029]
点击复制

基于视频分析的空管员违规行为识别方法*
分享到：

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

卷:
19
期数:
2023年1期
页码:
196-201
栏目:
职业安全卫生管理与技术
出版日期:
2023-01-31

文章信息/Info

Title:
Violations recognition method of air traffic controllers based on video analysis
文章编号:
1673-193X(2023)-01-0196-06
作者:
唐豪; 奉鑫鑫; 高曙; 罗帆; 揣明瑞
（1.武汉理工大学 计算机与人工智能学院，湖北 武汉 430070；
2.武汉理工大学 管理学院，湖北 武汉 430070)
Author(s):
TANG Hao;  FENG Xinxin;  GAO Shu;  LUO Fan;  CHUAI Mingrui
(1.School of Computer Science & Artificial Intelligence,Wuhan University of Technology,Wuhan Hubei 430070,China;
2.School of Management,Wuhan University of Technology,Wuhan Hubei 430070,China)
关键词:
违规行为; 深度学习; 视频监控; 空中交通管制员; 异常回归网络; ResNeXt网络
Keywords:
violations;  deep learning;  video surveillance;  air traffic controller;  abnormal regression network;  ResNeXt network
分类号:
X949
DOI:
10.11731/j.issn.1673-193x.2023.01.029
文献标志码:
A
摘要:
为利用视频数据对空管员违规行为进行智能化分析，降低不安全事件发生率，提出2阶段的违规行为识别模型（AR-ResNeXt），基于实地调研构建空管员视频数据集，利用最小化动态多实例学习损失函数和中心损失函数，获得违规行为检测的判别特征表示，结合异常回归网络和ResNeXt网络，完成对空管员违规行为的时序区间检测与动作分类。研究结果表明：AR-ResNeXt模型在自制数据集中，其帧级AUC达到82.9%，分类准确率达到87.8%，可准确识别空管员发生违规行为的时序区间并进行分类，研究结果可为保障空中交通安全奠定基础。
Abstract:
In order to use video data to intelligently analyze the violations of air traffic controllers and reduce the incidence of unsafe events,a two-stage violations recognition model (AR-ResNeXt) was proposed.Based on the video data set of air traffic controllers constructed by field investigation,the discriminant feature representation of violations detection was obtained by minimizing the dynamic multiple-instance learning loss function and the center loss function,and the time ordered interval detection and action classification of violations of air traffic controllers were completed by combining with the abnormal regression network and ResNext network.The results showed that the AR-ResNeXt model achieved a frame-level AUC of 86.50% and a classification accuracy of 87.8% on the self-made data set.It can accurately identify and classify the time ordered interval of violations of air traffic controllers,and lay a foundation for ensuring the air traffic safety.

参考文献/References:

［1］王华伟,吴海桥.航空安全工程［M］.北京:科学出版社,2014.
［2］温廷新,王贵通,孔祥博,等.基于迁移学习与残差网络的矿工不安全行为识别［J］.中国安全科学学报,2020,30(3):41-46. WEN Tingxin,WANG Guitong,KONG Xiangbo,et al.Identification of miners’ unsafe behaviors based on transfer learning and residual network ［J］.China Safety Science Journal,2020,30(3):41-46.
［3］张洪亮.基于大数据的煤矿违规行为分析识别系统研究［J］.煤矿安全,2022,53(5):133-137. ZHANG Hongliang.Research on mine violation analysis and identification system based on big data［J］.Safety in Coal Mines,2022,53(5):133-137.
［4］CHANG Y P,TU Z G,XIE W,et al.Clustering driven deep autoencoder for video anomaly detection［C］//European Conference on Computer Vision,Berlin，2020.
［5］ZHONG J X,LI N N,KONG W J,et al.Graph convolutional label noise cleaner:train a plug-and-play action classifier for anomaly detection［C］//Computer Vision and Pattern Recognition,Piscataway,2019.
［6］LIU W,LUO W X,LIAN D Z,et al.Future frame prediction for anomaly detection-a new baseline［C］ //Computer Vision and Pattern Recognition,Piscataway,2018.
［7］WAN B Y,FANG Y M,XIA X,et al.Weakly super-vised video anomaly detection via center-guided discriminative learning［C］//2020 IEEE International Conference on Multimedia and Expo,Piscataway,2020.
［8］XIE S,GIRSHICK R,DOLLR P,et al.Aggregated residual transformations for deep neural networks［C］//IEEE conference on Computer Vision and Pattern Recognition,Piscataway,2017.
［9］WANG L M,XIONG Y J,WANG Z,et al.Temporal segmentnetworks:towards good practices for deep action recognition［C］//European Conference on Computer Vision,Berlin,2016.
［10］崔铁军，王凌霄.YOLOv4目标检测算法在煤矿工人口罩佩戴监测工作中的应用研究［J］.中国安全生产科学技术，2021,17(10):66-71. CUI Tiejun，WANG Lingxiao.Research on application of YOLOv4 object detection algorithm in monitoring on masks wearing of coal miners［J］.Journal of Safety Science and Technology,2021,17(10):66-71.
［11］HE K M,ZHANG X Y,REN S Q,et al.Deep residual learning for image recognition［C］//Computer Vision and Pattern Recognition,Piscataway,2016.
［12］SZEGEDY C,IOFFE S,VANHOUCKE V,et al.Inception-v4,inception-resnet and the impact of residual connections on learning［C］//Thirty-first AAAI Conference on Artificial Intelligence,Menlo Park,2017.
［13］SULTANI W,CHEN C,SHAH M.Real-world anomaly detection in surveillance videos［C］//IEEE Conference on Computer Vision and Pattern Recognition,Piscataway,2018.
［14］TAN M,LE Q.Efficientnetv2:Smaller models and faster training［C］//International Conference on Machine Learning(ICML),New York,2021.

相似文献/References:

[1]杜珺,何家力.基于社会网络的危险品航空运输违规行为分析[J].中国安全生产科学技术,2018,14(7):73.[doi:10.11731/j.issn.1673-193x.2018.07.011]
　DU Jun,HE Jiali.Analysis on violation behavior in air transportation of dangerous goods based on social network[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(1):73.[doi:10.11731/j.issn.1673-193x.2018.07.011]
[2]涂思羽,彭平安,蒋元建.基于深度学习的井下环境异常工况智能识别技术研究[J].中国安全生产科学技术,2018,14(11):58.[doi:10.11731/j.issn.1673-193x.2018.11.009]
　TU Siyu,PENG Pingan,JIANG Yuanjian.Research on intelligent recognition technology of abnormal operating conditions in underground environment based on deep learning method[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(1):58.[doi:10.11731/j.issn.1673-193x.2018.11.009]
[3]刘欣宜,张宝峰,符烨,等.基于深度学习的污染场地作业人员着装规范性检测[J].中国安全生产科学技术,2020,16(7):169.[doi:10.11731/j.issn.1673-193x.2020.07.027]
　LIU Xinyi,ZHANG Baofeng,FU Ye,et al.Detection on normalization of operating personnel dressing at contaminated sites based on deep learning[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2020,16(1):169.[doi:10.11731/j.issn.1673-193x.2020.07.027]
[4]毕东月.基于深度学习的输煤皮带故障视觉检测方法研究[J].中国安全生产科学技术,2021,17(8):84.[doi:10.11731/j.issn.1673-193x.2021.08.013]
　BI Dongyue.Research on visual detection method for fault of coal conveyor belt based on deep learning[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2021,17(1):84.[doi:10.11731/j.issn.1673-193x.2021.08.013]
[5]崔铁军,王凌霄.YOLOv4目标检测算法在煤矿工人口罩佩戴监测工作中的应用研究*[J].中国安全生产科学技术,2021,17(10):66.[doi:10.11731/j.issn.1673-193x.2021.10.010]
　CUI Tiejun,WANG Lingxiao.Research on application of YOLOv4 object detection algorithm in monitoring on masks wearing of coal miners[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2021,17(1):66.[doi:10.11731/j.issn.1673-193x.2021.10.010]
[6]刘超,雷晨,李树刚,等.基于CNN-GRU的瓦斯浓度预测模型及应用*[J].中国安全生产科学技术,2022,18(9):62.[doi:10.11731/j.issn.1673-193x.2022.09.009]
　LIU Chao,LEI Chen,LI Shugang,et al.Prediction model of gas concentration based on CNN-GRU and its application[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2022,18(1):62.[doi:10.11731/j.issn.1673-193x.2022.09.009]
[7]曹亚利,李振雷,刘旭东,等.基于卷积神经网络的冲击地压预警方法研究*[J].中国安全生产科学技术,2022,18(10):101.[doi:10.11731/j.issn.1673-193x.2022.10.015]
　CAO Yali,LI Zhenlei,LIU Xudong,et al.Research on early-warning method of rockburst based on convolutional neural network[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2022,18(1):101.[doi:10.11731/j.issn.1673-193x.2022.10.015]
[8]李子奇,蒋柱虎,王力,等.基于深度学习的工程结构损伤识别研究进展[J].中国安全生产科学技术,2022,18(12):43.[doi:10.11731/j.issn.1673-193x.2022.12.006]
　LI Ziqi,JIANG Zhuhu,WANG Li,et al.Research progress in damage identification of engineering structure based on deep learning[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2022,18(1):43.[doi:10.11731/j.issn.1673-193x.2022.12.006]
[9]黄鹏,房志明,朱曼,等.基于SSD网络的电梯内电动自行车检测研究*[J].中国安全生产科学技术,2023,19(2):167.[doi:10.11731/j.issn.1673-193x.2023.02.023]
　HUANG Peng,FANG Zhiming,ZHU Man,et al.Research on detection of electric bicycles in elevator based on SSD network[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2023,19(1):167.[doi:10.11731/j.issn.1673-193x.2023.02.023]
[10]夏正洪,何琥,吴建军,等.基于深度学习的航空铆钉分类及异常情况检测*[J].中国安全生产科学技术,2023,19(6):199.[doi:10.11731/j.issn.1673-193x.2023.06.028]
　XIA Zhenghong,HE Hu,WU Jianjun,et al.Aviation rivet classification and abnormal situation detection based on deep learning[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2023,19(1):199.[doi:10.11731/j.issn.1673-193x.2023.06.028]

备注/Memo

备注/Memo:
收稿日期： 2022-03-04
* 基金项目： 国家自然科学基金项目(71271163)；国家教育部人文社科项目(18YJA630076)
作者简介： 唐豪，硕士研究生，主要研究方向为异常行为检测、行为识别。
通信作者： 高曙，博士，教授，主要研究方向为面向安全工程的智能计算、大数据分析。
更新日期/Last Update: 2023-02-14