基于改进YOLOv5s的航站楼应急状况检测算法

计算机科学 ›› 2026, Vol. 53 ›› Issue (6A): 250300174-7.doi: 10.11896/jsjkx.250300174

• 图像处理&多媒体技术 • 上一篇下一篇

基于改进YOLOv5s的航站楼应急状况检测算法

刘岱¹, 安鹏宇², 王凯²

1 中国民航大学工程技术训练中心天津 300300
2 中国民航大学电子信息与自动化学院天津 300300

出版日期:2026-06-16 发布日期:2026-06-12
通讯作者: 王凯(k-wang@cauc.edu.cn)
作者简介:(dliu@cauc.edu.cn)
基金资助:
国家重点研发计划(2023YFB4302901)

Improved YOLOv5s-based Algorithm for Emergency Situation Detection in Airport Terminals

LIU Dai¹, AN Pengyu², WANG Kai²

1 Engineering Techniques Training Center, Civil Aviation University of China,Tianjing 300300,China
2 School of Electronic Information and Automation,Civil Aviation University of China,Tianjing 300300,China

Published:2026-06-16 Online:2026-06-12
About author:LIU Dai,born in 1981,lecturer,master.His main research interest is aircraft system simulation technology.
WANG Kai,born in 1982,associate professor,master.His main research interests include avionics equipment testing and fault diagnosis,virtual simulation of airborne systems.
Supported by:
National Key R&D Program of China(2023YFB4302901).

摘要/Abstract

摘要： 针对航站楼对较高应急响应能力的需求,对现有的YOLOv5s目标检测模型进行改进优化,以适应航站楼在复杂环境下的应急状况检测。文中的检测目标有3类:火焰、烟雾和人员摔倒。该研究对传统YOLOv5s模型进行了3项改进:引入MPDIoU损失函数、引入softNMS非极大值抑制算法、在小目标层添加BiFormer注意力机制。该研究使用消融实验和对比实验证明了改进的有效性,实验结果表明,改进后的模型在使用自建数据集进行训练后表现出了优秀的性能,在mAP@0.5和mAP@0.5:0.95这两个平均精度指标上获得了显著提升,分别达到了93.1%和63.5%,相比YOLOv5s原模型分别提升了1.7%和4.2%,相比最新模型YOLOv11分别提升了1.1%和5%。改进后的模型在进行航站楼视频流检测时表现出较好的实时性能,并且模型权重较小使得部署比较容易,满足了航站楼应急状况检测的需求,具有较高的应用价值。

关键词: 目标检测, 航站楼安全, MPDIoU, softNMS, BiFormer

Abstract: Based on the response urgency requirements of terminals on emergency calls,this article optimizes the previous YOLOv5s object detection model to improve terminal situation emergency calls.There are three types of detection targets in the study:flames,smoke,and people's falls.Specifically,it improves the model by using MPDIoU as the loss function instead of original one,replacing NMS algorithm with softNMS algorithm and integrating the BiFormer attention mechanism into the small target detection layer.The study employes ablation experiments and comparative trials to validate the effectiveness of the improvements.Experimental results demonstrate that the enhanced model exhibits superior performance after training on a custom-built dataset,achieving significant improvements in the average precision metrics of mAP@0.5 and mAP@0.5:0.95,reaching 93.1% and 63.5% respectively.Compared to the original YOLOv5s model,these metrics increase by 1.7% and 4.2%,while outperforming the latest YOLOv11 model by 1.1% and 5% in the respective metrics.The better model works well with real-time video feeds from the airport terminals.It fulfills the requirements for incident detection and has high prospect of application at this kind of critical places.

Key words: Object detection, Terminal security, MPDIoU, softNMS, BiFormer

中图分类号:

TP391.41

刘岱, 安鹏宇, 王凯. 基于改进YOLOv5s的航站楼应急状况检测算法[J]. 计算机科学, 2026, 53(6A): 250300174-7. https://doi.org/10.11896/jsjkx.250300174

LIU Dai, AN Pengyu, WANG Kai. Improved YOLOv5s-based Algorithm for Emergency Situation Detection in Airport Terminals[J]. Computer Science, 2026, 53(6A): 250300174-7. https://doi.org/10.11896/jsjkx.250300174

参考文献

[1] YE L,QI Y Z,SHI J X,et al.A Review of Human Fall Detec-tion Technology [J].Electronic Test,2020(2):50-51,65.
[2] LIU Y,DENG Y,SU J,et al.Multiple Input Branches ShiftGraph Convolutional Network with Drop Edge for Skeleton-Based Action Recognition[C]//International Conference on Image Analysis and Processing Springer,2022.
[3] ABDO H,AMIN K M,HAMAD A M.Fall detection based on retinanet and mobilenet convolutional neural networks[C]//15th International Conference on Computer Engineering and Systems(ICCES 2020).IEEE,2020:1-7.
[4] LIN T Y,GOYAL P,GIRSHICK R,et al.Focal loss for dense object detection [J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2017,99:2999-3007.
[5] ZHU N.Elderly Fall Detection Based on Improved SSD Algorithm [D].North China University of Technology,2022.
[6] JIANG Y Y.Design and Implementation of a Home Fall Detection System for the Elderly Based on YOLOv3 [D].Nanjing:Nanjing University of Posts and Telecommunications,2021.
[7] CAO J R,LYU J J,WU X Y,et al.Fall Detection Algorithm Integrating Motion Features and Deep Learning [J].Journal of Computer Applications,2021,41(2):583-589.
[8] WANG Y,HUA C,DING W,et al.Real-time detection of flame and smoke using an improved YOLOv4 network[J].Signal,Image and Video Processing,2022,16(4):1109-1116.
[9] WU Z,XUE R,LI H.Real-time video fire detection via modified YOLOv5 network model[J].Fire Technology,2022,58(4):2377-2403.
[10] WANG A,LIANG G,WANG X,et al.Application of theYOLOv6 combining CBAM and CIoU in forest fire and smoke detection[J].Forests,2023,14(11):2261.
[11] LI Y,ZHANG W,LIU Y,et al.An efficient fire and smoke detection algorithm based on an end-to-end structured network[J].Engineering Applications of Artificial Intelligence,2022,116:105492.
[12] HU X,LI M,ZHANG D W.Flamenet:a lightweight convolu-tional neural network for flame detection and localisation[J].International Journal of Vehicle Design,2023,91(3):87-106.
[13] ZOU H J,JIAO L B,MENG L,et al.Flame and Smoke Detection Based on CG-YOLO [J].Computer & Digital Engineering,2022,50(1):206-212.
[14] REN S Q,HE K M,GIRSHICK R,et al.Faster R-CNN:Towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,39(6):1137-1149.
[15] LIU W,ANGUELOV D,EERHAN D,et al.SSD:singleshotmultibox detector[C]//European Conference on Computer Vision.Cham:Springer,2016:21-37.
[16] LIU S,HUANG D.Receptive field block net for accurate and fast object detection[C]//Proceedings of the European Conference on Computer Vision(ECCV).2018:385-400.
[17] ZHU L,WANG X,KE Z,et al.BiFormer:Vision Transformer with Bi-Level Routing Attention[J].arXiv:2303.08810.
[18] MA S,XU Y.MPDIoU:A Loss for Efficient and AccurateBounding Box Regression [J].arXiv:2307.07662,2023.
[19] BODLA N,SINGH B,CHELLAPPA R,et al.Improving Object Detection With One Line of Code[C]//Proceedings of the lEEE Interational Conference on Computer Vision.2017:5561-5569.
[20] ZHENG Z,WANG P,LIU W,et al.Distance-IoU loss:faster and better learning for bounding box regression[C]//Proceedings of the AAAI Conference on Artificial Intel Ligence.2020:12993-13000.
[21] ZHANG Y F,REN W,ZHANG Z,et al.Focal and effi cient IOU loss for accurate bounding box regression[J].Neurocomputing,2022,506:146-157.

相关文章 0

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

本文评价

推荐阅读 0

No Suggested Reading articles found!