| 56 | 0 | 835 |
| 下载次数 | 被引频次 | 阅读次数 |
提出快速检测无人机影像窨井盖目标位置的YOLO-HR模型，该模型以YOLOv11模型结构为基础，采用混合域通道注意力模块(Hybrid Domain Channel Attention, HDCA)增强模型主干层的多尺度特征融合与提取能力，并在检测头部分引入残差注意力模块(Residual Attention Block, RAB)用于提升模型检测头的特征捕捉能力。实验结果表明：本文模型相对于YOLOv11基线模型平均检测精度较高，在精确率、召回率、F1、AP@0.5及AP@0.5∶0.95指标上分别提升了0.7%、17.7%、12.4%、5.3%、4.4%;跨区域泛化实验显示，本文模型在未见场景中仍具有稳定的检测性能与良好的适应性。
Abstract：To facilitate rapid detection of manhole cover targets, this paper proposes a YOLO-HR model for locating manhole covers in UAV images.First, a semi-automatic method for constructing a UAV-based manhole cover dataset is developed.Next, the YOLO-HR model is introduced to achieve fast and accurate manhole cover detection.Based on the YOLOv11 architecture, the model incorporates a hybrid domain channel attention(HDCA) module to enhance multi-scale feature fusion and extraction in the backbone network, and integrates a residual attention block(RAB) into the detection head to strengthen feature representation capability.Finally, experiments are conducted to validate the effectiveness and advantages of the proposed model.Experimental results demonstrate that the proposed model achieves high precision, recall, mean average precision(mAP),and F1-score.Compared with the baseline YOLOv11 model, it improves precision, recall, F1-score, AP@0.5,and AP@0.5∶0.95 by 0.7%,17.7%,12.4%,5.3%,and 4.4%,respectively.Cross-region generalization experiments further show that the proposed model maintains stable detection performance and strong adaptability in unseen scenarios.
[1] 龚敏霞，袁赛，储征伟，等.顾及多空间相似性的地下管线数据匹配[J].测绘学报，2015,44(12):1392-1400.
[2] 查旭东，眭子凡，张浚轩，等.道路窨井盖-井周路面的病害处治与智慧检测监管综述[J].中国公路学报，2024,37(12):357-380.
[3] Ravi Kumar K,Vijaya Lakshmi K,Rohin Kumar G,et al.Smart manhole monitoring system[C]//2023 5th International Conference on Smart Systems and Inventive Technology (ICSSIT).Tirunelveli:IEEE,2023:475-481.
[4] 郑丰收，周文，宋永明.城市井盖智能化管理[J].测绘通报，2013(S2):55-58.
[5] 高铁军，吴立新.论城市管网智慧管理研究范畴与关键技术[J].地理与地理信息科学，2011,27(4):19-23.
[6] 顾敏.管好小小井盖，守好“脚下安全”[N].新华日报，2024-10-18(3).
[7] 杨永春，菅煜婷.人工智能时代城市地理学发展的变革与挑战[J].地理学报，2024,79(10):2425-2441.
[8] 张新长，华淑贞，齐霁，等.新型智慧城市建设与展望：基于AI的大数据、大模型与大算力[J].地球信息科学学报，2024,26(4):779-789.
[9] Zhou Baoding,Zhao Wenjian,Guo Wenhao,et al.Smartphone-based road manhole cover detection and classification[J].Automation in Construction,2022,140:104344.DOI:10.1016/j.autcon.2022.104344.
[10] 周长江，张晨辉.新型测绘技术在城市部件普查中的应用[J].测绘通报，2023(12):147-152.
[11] 高笔清，杨小东，张旭冬.车载移动测量系统在城市部件普查中的应用[J].测绘通报，2020(S1):6-8.
[12] 杨蒙蒙，万幼川，刘先林，等.基于地面移动测量系统的井盖快速自动定位与提取方法的研究[J].中国激光，2018,45(8):128-135.
[13] Liu Haoting,Yan Beibei,Wang Wei,et al.Manhole cover detection from natural scene based on imaging environment perception[J].KSII Transactions on Internet and Information Systems,2019,13(10):5095-5111.
[14] Redmon J,Divvala S,Girshick R,et al.You only look once:unified,real-time object detection[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Las Vegas:IEEE,2016:779-788.
[15] Ren Shaoqing,He Kaiming,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,2017,39(6):1137-1149.
[16] Zhang Hang,Dong Zishuo,He Anzheng,et al.Efficient approach to automated pavement manhole cover detection with modified faster R-CNN[J].Intelligent Transportation Infrastructure,2022,1:liac006.DOI:10.1093/iti/liac006.
[17] Jin Lijun,Ding Wendi,Han Shijia,et al.A real-time edge inference method for insulator contamination detection with YOLOv11-ssL[J].IEEE Transactions on Instrumentation and Measurement,2025,74:1-15.
[18] Yang Li,Hao Zhongyu,Hu Bo,et al.Improved YOLOX-based detection of condition of road manhole covers[J].Frontiers in Built Environment,2024,10:1337984.DOI:10.3389/fbuil.2024.1337984.
[19] Liu Jing,Zhao Jianyong,Cao Yanyan,et al.Road manhole cover defect detection via multi-scale edge enhancement and feature aggregation pyramid[J].Scientific Reports,2025,15(1):10346.DOI:10.1038/s41598-025-95450-8.
[20] 孔天宇，戴激光.改进YOLOv5的路面井盖病害检测[J].遥感信息，2023,38(3):40-46.
[21] Zhang Zhengxin,Zhu Lixue.A review on unmanned aerial vehicle remote sensing:platforms,sensors,data processing methods,and applications[J].Drones,2023,7(6):398.DOI:10.3390/drones7060398.
[22] 程传祥，金飞，林雨准，等.应用多尺度融合策略和改进YOLOV5的道路病害无人机检测[J].地球信息科学学报，2024,26(8):1991-2007.
[23] 王畅，熊汉江，涂建光，等.无人机影像的松材线虫病半监督学习检测方法[J].武汉大学学报(信息科学版),2025,50(12):2560-2568.
[24] 杨梦圆，刘伟，尹鹏程，等.深度卷积网络支持下的遥感影像井盖部件检测[J].测绘通报，2019(8):78-81,87.
[25] Wang Dejiang,Huang Yuping.Manhole cover classification based on super-resolution reconstruction of unmanned aerial vehicle aerial imagery[J].Applied Sciences,2024,14(7):2769.DOI:10.3390/app14072769.
[26] 杨晋.Bjoern Altmann的新书：井盖之美[J].装饰，2023(10):10.
[27] Khanam R,Hussain M.YOLOv11:an overview of the key architectural enhancements[EB/OL].(2024-10-23)[2025-12-05].https://arxiv.org/abs/2410.17725.
[28] Myo N N,Boonkong A,Khampitak K,et al.A two-point association tracking system incorporated with YOLOv11 for real-time visual tracking of laparoscopic surgical instruments[J].IEEE Access,2025,13:12225-12238.
[29] Wu Wencong,Liu Shijie,Xia Yuelong,et al.Dual residual attention network for image denoising[J].Pattern Recognition,2024,149:110291.DOI:10.1016/j.patcog.2024.110291.
[30] Hu Jie,Shen Li,Sun Gang.Squeeze-and-excitation networks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Salt Lake City:IEEE,2018:7132-7141.
[31] Gao Zilin,Xie Jiangtao,Wang Qilong,et al.Global second-order pooling convolutional networks[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).Long Beach:IEEE,2019:3019-3028.
[32] Wang Qilong,Wu Banggu,Zhu Pengfei,et al.ECA-Net:efficient channel attention for deep convolutional neural networks[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).Seattle:IEEE,2020:11531-11539.
[33] Tan Chunhai,Chen Tao,Liu Jiayu,et al.Building extraction from unmanned aerial vehicle (UAV) data in a landslide-affected scattered mountainous area based on Res-Unet[J].Sustainability,2024,16(22):9791.DOI:10.3390/su16229791.
[34] Liu Hanqiang,Lu Sipei,Zhao Feng.MLP-Res-Unet:MLPs and residual blocks-based U-shaped network intervertebral disc segmentation of multi-modal MR spine images[J].Current Medical Imaging,2024,20:20.DOI:10.2174/1573405620666230417082855.
[35] Luo Jian,Zhang Yiying,Wu Yannian,et al.A multi-channel contrastive learning network based intrusion detection method[J].Electronics,2023,12(4):949.DOI:10.3390/electronics12040949.
[36] Ye Zhonglin,Tang Yanlong,Zhao Haixing,et al.Multi-channel high-order network representation learning research[J].Frontiers in Neurorobotics,2024,18:1340462.DOI:10.3389/fnbot.2024.1340462.
[37] He Kaiming,Zhang Xiangyu,Ren Shaoqing,et al.Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Las Vegas:IEEE,2016:770-778.
[38] Woo S,Park J,Lee J Y,et al.CBAM:convolutional block attention module[C]//Computer Vision-ECCV 2018.Cham:Springer,2018:3-19.
[39] Fu Chuan,Du Bo,Zhang Liangpei.Do we need learnable classifiers?A hyperspectral image classification algorithm based on attention-enhanced ResBlock-in-ResBlock and ETF classifier[J].IEEE Transactions on Geoscience and Remote Sensing,2024,62:1-13.
[40] Zheng Yuxuan,Li Jiaojiao,Li Yunsong,et al.Hyperspectral pansharpening using deep prior and dual attention residual network[J].IEEE Transactions on Geoscience and Remote Sensing,2020,58(11):8059-8076.
[41] Yaseen M.What is YOLOv8:an in-depth exploration of the internal features of the next-generation object detector[EB/OL].(2024-08-28)[2025-12-05].https://arxiv.org/abs/2408.15857.
[42] Tian Yunjie,Ye Qixiang,Doermann D.YOLOv12:attention-centric real-time object detectors[EB/OL].(2025-02-18)[2025-12-05].https://arxiv.org/abs/2502.12524.
[43] Lei Mengqi,Li Siqi,Wu Yihong,et al.YOLOv13:real-time object detection with hypergraph-enhanced adaptive visual perception[EB/OL].(2025-09-05)[2025-12-05].https://arxiv.org/abs/2506.17733.
[44] Zhang Yin,Ye Mu,Zhu Guiyi,et al.FFCA-YOLO for small object detection in remote sensing images[J].IEEE Transactions on Geoscience and Remote Sensing,2024,62:1-15.
[45] Wang Shuo,Xia Chunlong,Lv Feng,et al.RT-DETRv3:real-time end-to-end object detection with hierarchical dense positive supervision[EB/OL].(2024-09-13)[2025-12-05].https://arxiv.org/html/2409.08475v1.
基本信息:
中图分类号:TU990.3;TP183;TP391.41
引用信息:
[1]徐青,朱新铭,刘怡,等.YOLO-HR:基于混合域通道注意力与残差注意力的窨井盖检测算法[J].地理与地理信息科学,2026,42(02):43-50+90.
基金信息:
国家自然科学基金项目(42101455)
2026-03-25
2026-03-25