[1]索智文,丁剑明,屈波,等.基于深度学习的矿井视频流异常检测算法研究[J].中国安全生产科学技术,2025,21(3):133-140.[doi:10.11731/j.issn.1673-193x.2025.03.017]
SUO Zhiwen,DING Jianming,QU Bo,et al.Research on anomaly detection algorithm of video stream in mine based on deep learning[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2025,21(3):133-140.[doi:10.11731/j.issn.1673-193x.2025.03.017]
点击复制
基于深度学习的矿井视频流异常检测算法研究
SUO Zhiwen,DING Jianming,QU Bo,et al.Research on anomaly detection algorithm of video stream in mine based on deep learning[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2025,21(3):133-140.[doi:10.11731/j.issn.1673-193x.2025.03.017]
《中国安全生产科学技术》[ISSN:1673-193X/CN:11-5335/TB]
- 卷:
- 21
- 期数:
- 2025年3期
- 页码:
- 133-140
- 栏目:
- 职业安全卫生管理与技术
- 出版日期:
- 2025-03-31
文章信息/Info
- Title:
- Research on anomaly detection algorithm of video stream in mine based on deep learning
- 文章编号:
- 1673-193X(2025)-03-0133-08
- Keywords:
- YOLOv4 algorithm; video surveillance; video stream anomaly detection; MAP; intelligent mining
- 分类号:
- X936
- 文献标志码:
- A
- 摘要:
- 为了探究矿井复杂环境中视频流检测精度问题，提出1种基于YOLOv4深度优化的复杂环境视频流异常检测算法，增设SE模块提升特征提取效率，改进SPP、PANet模块优化异常检测能力；提取矿井现场真实数据，对数据集中4 500多张异常行为进行模型训练，采用深度优化的YOLOv4算法进行识别，标注出视频异常行为。研究结果表明：相较于传统的YOLOv4算法，深度优化后的模型平均精确率均值(MAP)为98.02%，MAP提升16.6百分点，每秒传输帧数(FPS)提高至28.56。研究结果可为优化矿井复杂环境下视频流检测精度提供思路和方法。
- Abstract:
- To investigate the issue of video stream detection accuracy in complex environment of mine,an anomaly detection algorithm of video stream in complex environment based on deeply optimized YOLOv4 was proposed.A SE module was added to enhance the feature extraction efficiency,and the SPP and PANet modules were improved to optimize the anomaly detection capability.The on-site real data of mine was extracted,and over 4 500 abnormal behavior in the dataset were used for model training.The deeply optimized YOLOv4 algorithm was employed for the identification,and the abnormal behavior in the video was annotated.The results show that tcomparing with the traditional YOLOv4 algorithm, the mean Average Precision (MAP) of the deeply optimized model was 98.02%, increasing by 16.6 percentage points, and the Frames Per Second (FPS) of transmission was increased to 28.56.The research findings can provide insights and methods for optimizing the accuracy of video stream detection in complex environment of mines.
参考文献/References:
［1］梁燕华,芦君珂,田训鹏.基于改进分布估计的算法在井下监控系统中的应用研究［J］.中国矿业,2023,32(12):92-99.
LIANG Yanhua,LU Junke,TIAN Xunpeng.Research on the application of an improved distribution estimation algorithm in underground monitoring systems ［J］.China Mining Magazine,2023,32(12):92-99.
［2］罗庆中,陈志军.基于ZigBee的矿井安全监测系统设计［J］.内蒙古煤炭经济,2024(10):108-110.
LUO Qingzhong,CHEN Zhijun.Design ofmine safety monitoring system based on ZigBee ［J］.Inner Mongolia Coal Economy,2024(10):108-110.
［3］范颖晨,褚治广,李启龙.基于视频流分析的重点区域异常行为监测平台［J］.辽宁工业大学学报(自然科学版),2024,44(2):109-114.
FAN Yingchen,CHU Zhiguang,LI Qilong.Abnormal behavior monitoring platform in key regions based on video stream analysis ［J］.Journal of Liaoning University of Technology (Natural Science Edition),2024,44 (2):109-114.
［4］孙晓.基于深度学习的异常行为检测与识别方法研究［D］.北京：北京交通大学,2023.
［5］梁睿琳,王锐,郭迎.基于视频时空关系的高速公路异常停车检测［J］.计算机应用研究,2022,39(6):1916-1920.
LIANG Ruilin,WANG Rui,GUO Ying.Freeway abnormal parking detection based on video spatiotemporal relationship ［J］.Journal of Applied Research of Computers,2022,39(6):1916-1920.
［6］陈翰林.基于YOLOv4-tiny的煤矿职工井下违章行为识别的研究［D］.淮南：安徽理工大学,2021.
［7］苏晨,郝乐,杨若楠.基于视频流的井下人数统计方法研究与应用［J］.煤炭技术,2019,38(12):164-167.
SU Chen,HAO Le,YANG Ruonan.Research and application of downhole man count method based on video stream ［J］.Coal Technology,2019,38(12):164-167.
［8］赵仁凤.视频监控中人体异常行为识别［J］.宿州学院学报,2018,33(11):111-115.
ZHAO Renfeng.Human abnormal behavior recognition in video surveillance ［J］.Journal of Suzhou University,2018,33(11):111-115.
［9］原磊明,王海斌,刘旭飞,等.矿井人身安全视频防护系统设计［J］.煤炭技术,2018,37(11):373-375.
YUAN Leiming,WANG Haibin,LIU Xufei,et al.Design of video protection system for mine personal safety ［J］.Coal Technology,2018,37(11):373-375.
［10］ASAD M,JIANG H,YANG J,et al.Multi-stream 3D latent feature clustering for abnormality detection in videos ［J］.Applied Intelligence,2021,52(1):1-18.
［11］OZAGA A,MURAT E.A data stream-based approach for anomaly detection in surveillance videos ［J］.Multimedia Tools and Applications,2024,83(21):60213-60241.
［12］ZHAO Y X,MANK L.A novel two-stream structure for video anomaly detection in smart city management ［J］.The Journal of Supercomputing,2022,78:3940-3954.
［13］LI S,WANG Z G,ZHANG Y J,et al.A feature-trajectory-smoothed high-speed model for video anomaly detection ［J］.Sensors,2023,23(3):1612-1612.
［14］ZHENG T X,JIANG M Z,LI Y F,et al.Research on tomato detection in natural environment based on RC-YOLOv4 ［J］.Computers and Electronics in Agriculture,2022,198:107029.
［15］LI Q,ZHAO F,XU Z P,et al.Improved YOLOv4 algorithm for safety management of on-site power system work ［J］.Energy Reports,2022,8(13):739-746.
［16］JIANG S,YIN S R,LUO T H,et al. Multi-target vehicle detection algorithm based on improved YOLOv4［J］.Computer Engineering and Design,2024,45(4):1181-1188.
［17］赵瑞.视频传输算法在浮选泡沫图像分析系统中的应用［J］.中国矿业,2022,31(增刊1):315-320.
ZHAO Rui.Research onthe application of video transmission algorithms in the flotation foam image analysis system ［J］.China Mining Magazine,2022,31(Supplement 1):315-320.
［18］白莹.基于视频智能识别的带式运输机安全检测系统［J］.现代矿业,2023,39(6):59-62.
BAI Ying.Safety detection system of beltconveyer based on video intelligent recognition ［J］.Modern Mining,2023,39(6):59-62.
［19］李飞,胡坤,张勇,等.基于混合域注意力YOLOv4的输送带纵向撕裂多维度检测［J］.浙江大学学报(工学版),2022,56(11):2156-2167.
LI Fei,HU Kun,ZHANG Yong,et al.Multi-dimensional detection of longitudinal tearing of conveyor belt based on mixed domain attention YOLOv4 ［J］.Journal of Zhejiang University (Engineering and Technology),2022,56 (11):2156-2167.
［20］郭永存,杨豚,王爽.基于改进YOLOv4-Tiny的矿井电机车多目标实时检测［J］.工程科学与技术,2023,55(5):232-241.
GUO Yongcun,YANG Tun,WANG Shuang.Multi-object real-time detection of mine electric locomotive based onimproved YOLOv4-Tiny ［J］.Advanced Engineering Sciences,2023,55(5): 232-241.
［21］阮顺领,董莉娟,卢才武,等.基于RCR＿YOLOv4的矿井巷道红外障碍检测研究［J］.黄金科学技术,2022,30(4):603-611.
RUAN Shunling,DONG Lijuan,LU Caiwu,et al.Research on infrared obstacle detection of mine roadway based on RCR_YOLOv4 ［J］.Gold Science and Technology,2022,30(4):603-611.
［22］边铁山.基于SE-YOLOv5模型皮带异物检测算法研究 ［J］.中国矿业,2024,33(7):127-134.
BIAN Tieshan.Study onforeign object detection algorithm for conveyor belts based on the SE-YOLOv5 Model ［J］.China Mining Magazine,2024,33(7):127-134.
［23］李晶,洪武,张文亮,等.基于YOLOv5算法的智能剥锌机预开口识别技术研究［J］.中国矿业,2024,33(增刊1):258-262,267.
LI Jing,HONG Wu,ZHANG Wenliang,et al.Investigation intopre-cut identification technology for intelligent zinc stripping machines using the YOLOv5 algorithm ［J］.China Mining Magazine,2024,33(Supplement 1):258-262,267.
［24］杨建伟,涂兴子,梅峰漳,等.基于深度学习优化YOLOV3算法的芳纶带检测算法研究［J］.中国矿业,2020,29(4):67-72.
YANG Jianwei,TU Xingzi,MEI Fengzhang,et al.Research on the optimization of YOLOV3 algorithm for aramid belt detection using deep learning techniques ［J］.China Mining Magazine,2020,29(4):67-72.
相似文献/References:
[1]康荣学.基于GIS的重大危险源安全监测预警系统研究与开发[J].中国安全生产科学技术,2010,6(3):110.
KANG Rong-xue.Research and development of safety minitoring and early-warning system based on GIS for the major hazards[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2010,6(3):110.
[2]章耿勇.超高压电网应急指挥信息平台的设计与应用[J].中国安全生产科学技术,2013,9(3):129.[doi:10.11731/j.issn.1673-193x.2013.03.23]
ZHANG Geng yong.Design and application of emergency command information platform for EHV power grid[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2013,9(3):129.[doi:10.11731/j.issn.1673-193x.2013.03.23]
[3]康荣学,吴宗之.烟花爆竹企业安全监控预警系统研究*[J].中国安全生产科学技术,2009,5(6):21.
KANG Rong xue,WU Zong zhi.Research of safety monitoring system for fireworks and crackers enterprise[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2009,5(3):21.
[4]郭志宇,郭安宁,白雪见,等.基于视频监控系统的突发灾害应急评估技术可行性研究[J].中国安全生产科学技术,2017,13(5):122.[doi:10.11731/j.issn.1673-193x.2017.05.020]
GUO Zhiyu,GUO Anning,BAI Xuejian,et al.Feasibility study on emergency assessment technology of sudden disaster based on video surveillance system[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2017,13(3):122.[doi:10.11731/j.issn.1673-193x.2017.05.020]
[5]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.[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(3):196.[doi:10.11731/j.issn.1673-193x.2023.01.029]
备注/Memo
- 备注/Memo:
-
收稿日期： 2024-11-05
作者简介： 索智文，硕士，高级工程师，主要研究方向为煤矿智能化建设。
