WEI Chunmei,MA Shangchang,LU Huiguo,et al.Research on Equipment Monitoring Technology of Meteorological Observation Field based on Video Recognition[J].Journal of Chengdu University of Information Technology,2023,38(02):129-135.[doi:10.16836/j.cnki.jcuit.2023.02.001]
基于视频识别的气象观测场设备监控技术研究
- Title:
- Research on Equipment Monitoring Technology of Meteorological Observation Field based on Video Recognition
- 文章编号:
- 2096-1618(2023)02-0129-07
- Keywords:
- target detection; SSD; Yolov5; data enhancement; video recognition
- 分类号:
- TP391
- 文献标志码:
- A
- 摘要:
- 针对观测场长期无人监管,存在设备被人破坏和偷盗,且目前还没有对观测场设备进行实时监控的有效算法和系统等问题,比较目前典型的目标检测算法,将其应用于观测场设备的检测和识别。根据研究目标搭建数据获取平台,研究数据预处理算法,比较经典目标检测算法SSD和Yolov5两种模型在气象观测设备识别中的效果。两种模型实验对比结果表明,SSD的识别准确率为92.09%,但训练模型速度较快; Yolov5的识别准确率为95.82%,收敛很快,且识别结果较佳。
- Abstract:
- In view of the long-term unsupervised observation field, the equipment is damaged and stolen, and there is no effective algorithm and system for real-time monitoring of observation field equipment,the current typical target detection algorithms are compared and applied to the detection and identification of observation field equipment.The data acquisition platform is built according to the research objectives,the data preprocessing algorithm is studied,and the effects of two classical target detection algorithms SSD and Yolov5 model in the recognition of meteorological observation equipment are compared.The experimental results show that the detection speed of SSD is faster,but the accuracy is only 92.09%,the accuracy of Yolov5 is 95.82% and the convergence is fast.
参考文献/References:
[1] GB 31221-2014,气象探测环境保护规范 地面气象观测站[S].
[2] 杨涛,张常亮,朱墨.地面气象观测场实景监控系统设计[J].成都信息工程学院学报,2013,28(4):336-341.
[3] 索子恒.图像特征检测与特征提取综述[J].产业创新研究,2022(4):33-35.
[4] 张静,农昌瑞,杨智勇.基于卷积神经网络的目标检测算法综述[J/OL].兵器装备工程学报:1-12[2022-0602].http://kns.cnki.net/kcms/detail/50.1213.TJ.20220427.1656.035.html.
[5] 寇大磊,权冀,张仲伟.基于深度学习的目标检测框架进展研究[J].计算机工程与应用,2019,55(11).
[6] Wei Liu,Dragomir Anguelov,Dumitru Erhan,et al.SSD:Single Shot MultiBox Detector[J].CoRR,2015,abs/1512:02325.
[7] Jeong J,Park H,Kwak N.Enhancement of SSD by concatenating feature maps for object detection[J].British Machine Vision Conference 2017,1705:09587.
[8] Redmon J,Divvala S,Girshick R,et al.You Only Look Once:Unified,Real-Time Object Detection[J].Computer Vision & Pattern Recognition,2016.
[9] Redmon J,Farhadi A.YOLO9000:Better,Faster,Stronger[J].IEEE Conference on Computer Vision & Pattern Recognition,2017:6517-6525.
[10] Redmon J,Farhadi A.YOLOv3:An Incremental Improvement[J].arXiv e-prints,2018.
[11] Bochkovskiy A,Wang C Y,Liao H.YOLOv4:Optimal Speed and Accuracy of Object Detection[J].2020,2004:10934.
[12] 杨永辉,刘昌平,黄磊.图像和视频分析在电力设备监控系统中的应用[J].计算机应用,2010,30(S1):281-284.
[13] 朱晓慧,钱丽萍,傅伟.图像数据增强技术研究综述[J].软件导刊,2021,20(5):230-236.
[14] 谭红臣,李淑华,刘彬,等.特征增强的SSD算法及其在目标检测中的应用[J].计算机辅助设计与图形学学报,2019,31(4):573-579.
[15] 谈世磊,别雄波,卢功林,等.基于YOLOv5网络模型的人员口罩佩戴实时检测[J].激光杂志,2021,42(2):147-150.
相似文献/References:
[1]李 静,鲜 林,王海江.基于YOLOv3的船只检测算法研究[J].成都信息工程大学学报,2023,38(01):37.[doi:10.16836/j.cnki.jcuit.2023.01.006]
LI Jing,XIAN Lin,WANG Haijiang.Research on Ship Detection Algorithm based on YOLOv3[J].Journal of Chengdu University of Information Technology,2023,38(02):37.[doi:10.16836/j.cnki.jcuit.2023.01.006]
[2]孙光灵,周云龙.自注意力结合上下文解耦的交通车辆检测[J].成都信息工程大学学报,2024,39(04):422.[doi:10.16836/j.cnki.jcuit.2024.04.005]
SUN Guangling,ZHOU Yunlong.Traffic Vehicle Detection based on Self-Attention Combined with Context Decoupling[J].Journal of Chengdu University of Information Technology,2024,39(02):422.[doi:10.16836/j.cnki.jcuit.2024.04.005]
备注/Memo
收稿日期:2022-06-16
基金项目:国家自然科学基金资助项目(42075129)