[1]刘润杰,许慧娜,胡　宇,等.基于改进YOLOv8的遥感影像变电站目标识别[J].郑州大学学报(工学版),2026,47(01):33-40.[doi:10.13705/j.issn.1671-6833.2025.04.022]
　LIU Runjie,XU Huina,HU Yu,et al.Remote Sensing Image Substation Target Recognition Based on Improved YOLOv8[J].Journal of Zhengzhou University (Engineering Science),2026,47(01):33-40.[doi:10.13705/j.issn.1671-6833.2025.04.022]
点击复制

基于改进YOLOv8的遥感影像变电站目标识别()
分享到：

《郑州大学学报(工学版)》[ISSN:1671-6833/CN:41-1339/T]

卷:
47
期数:
2026年01期
页码:
33-40
栏目:
出版日期:
2026-01-06

文章信息/Info

Title:
Remote Sensing Image Substation Target Recognition Based on Improved YOLOv8
文章编号:
1671-6833(2026)01-0033-08
作者:
刘润杰1; 2;  许慧娜1; 2;  胡　宇1; 2;  王　一1;  谢国钧1; 3
1.郑州大学 国家超级计算郑州中心,河南 郑州 450001;2.郑州大学 计算机与人工智能学院,河南 郑州 450001;3.中科星图金能(南京)科技有限公司,江苏 南京 211100
Author(s):
LIU Runjie1; 2;  XU Huina1; 2;  HU Yu1; 2;  WANG Yi1 ;  XIE Guojun1; 3
1.National Supercomputing Center in Zhengzhou, Zhengzhou University, Zhengzhou 450001, China; 2.School of Computer Science and Artificial Intelligence, Zhengzhou University, Zhengzhou 450001, China; 3.Zhongke Xingtu Jinneng (Nanjing) Technology Co., Ltd., Nanjing 211100, China
关键词:
YOLOv8;  遥感影像;  目标检测;  变电站;  注意力机制
Keywords:
YOLOv8;  remote sensing image;  object detection;  substation;  attention mechanism
分类号:
TP391TP751TM63
DOI:
10.13705/j.issn.1671-6833.2025.04.022
文献标志码:
A
摘要:
针对现有研究多集中于变电站局部结构检测而缺乏大区域快速发现与动态监测的问题,通过高分辨率卫星影像实现变电站的高效识别,提升电网安全隐患排查能力。首先构建了基于高分辨率光学卫星影像的变电站目标检测样本库;随后提出改进的YOLOv8算法,在骨干网络中嵌入SimAM轻量级注意力模块以增强细部特征聚焦能力,并将颈部结构替换为Efficient-RepGFPN,结合DySample动态上采样模块设计新型颈部结构GDFPN,以解决多层级特征语义错位问题。实验结果表明:改进方法优于主流检测算法,mAP75和mAP50-95分别提升至96.8%和87.1%,验证了其在变电站检测任务中的优越性。所提出的改进YOLOv8方法可有效支持大区域变电站的快速发现与动态监测,为电网安全管理提供了可靠的技术支撑。
Abstract:
Aiming at the limitation in existing studies focused on the detection of substation local structures, such as lacking methods for rapid discovery and dynamic monitoring over large areas, the capability of identifying potential safety hazards in power grids was enhanced through high-resolution satellite imagery. Firstly, a substation object detection dataset based on high-resolution optical satellite imagery was constructed. Subsequently, an improved YOLOv8 algorithm was proposed, embedding the SimAM lightweight attention module into the backbone network to enhance the ability to focus on detailed features, and replacing the neck with an Efficient-RepGFPN, combined with a DySample dynamic upsampling module to design a novel neck named GDFPN, addressing issues of multilevel feature semantic misalignment. Experimental results demonstrated that the improved method outperformed mainstream detection algorithms, with mAP75 and mAP50-95 increasing to 96.8% and 87.1%, respectively, confirming its superiority in substation detection tasks. The improved YOLOv8 approach proposed could effectively support the rapid discovery and dynamic monitoring of substations over large areas, providing reliable technical support for the safety management of power grids.

参考文献/References:

[1]REN S Q, HE K M, GIRSHICK R, et al. Faster RCNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 11371149.

[2]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). Piscataway: IEEE, 2016: 779788.
[3]REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]∥2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE, 2017: 6517-6525.
[4]REDMON J, FARHADI A. YOLOv3: an incremental improvement[EB/OL]. (2018-04-08)[2025-03-29]. https:∥arxiv.org/abs/1804.02767.
[5]BOCHKOVSKIY A, WANG C Y, LIAO H M. YOLOv4: optimal speed and accuracy of object detection[EB/OL]. (2020-04-23)[2025-03-29]. https:∥arxiv. org/abs/2004.10934.
[6]WANG C Y, BOCHKOVSKIY A, LIAO H M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for realtime object detectors[C]∥2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE, 2023: 7464-7475.
[7]LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot MultiBox detector[C]∥Computer Vision-ECCV 2016. Cham: Springer, 2016: 21-37.
[8]SUN H B, SHEN Q C, KE H C, et al. Power transmission lines foreign object intrusion detection method for drone aerial images based on improved YOLOv8 network[J]. Drones, 2024, 8(8): 346. 
[9]SHI J Y, BAI Y H, ZHOU J, et al. Multi-crop navigation line extraction based on improved YOLO-v8 and threshold-DBSCAN under complex agricultural environments[J]. Agriculture, 2024, 14(1): 45. 
[10] XUE F C, CHENG Y N, SHEN Y F, et al. Urban flood inundation area detection using YOLOv8 model[J]. Water Resources Management, 2025,39: 5443-5460.
[11]邝先验, 程福军, 吴翠琴, 等. 基于改进YOLOv7-tiny的高效轻量遥感图像目标检测方法[J]. 电子测量与仪器学报, 2024,38(7): 22-33.
KUANG X Y, CHENG F J, WU C Q, et al. Efficient and lightweight target detection method for remote sensing images based on improved YOLOv7-tiny[J]. Journal of Electronic Measurement and Instrument, 2024,38(7): 22-33.
[12]魏明军, 王镆涵, 刘亚志, 等. 基于特征融合和混合注意力的小目标检测[J]. 郑州大学学报(工学版), 2024, 45(3): 72-79.
WEI M J, WANG M H, LIU Y Z, et al. Small object detection based on feature fusion and mixed attention[J]. Journal of Zhengzhou University (Engineering Science), 2024, 45(3): 72-79.
[13]鲁玲,李明良,熊威,等. 基于改进的YOLOv8变电设备红外图像检测[J].红外技术,2025,47(9):1135-1141.
LU L, LI M L, XIONG W, et al. Infrared image detection of substation equipment based on improved YOLOv8[J]. Infrared Technology,2025,47(9):1135-1141.
[14]李斌, 李亚霖, 朱新山, 等. 基于注意力机制与特征平衡的变电站多目标检测[J]. 电网技术, 2022, 46(6): 2122-2132.
LI B, LI Y L, ZHU X S, et al. Multi-target detection in substation scence based on attention mechanism and feature balance[J]. Power System Technology, 2022, 46(6): 2122-2132.
[15] PENG J, LIU Y B, MA Y, et al. Deep transfer learning for power substation recognition with Google Earth[C]∥2018 15th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP). Piscataway: IEEE, 2018: 54-57.
[16] PÉREZ-HERNÁNDEZ F, RODRÍGUEZ-ORTEGA J, BENHAMMOU Y, et al. CI-dataset and DetDSCI methodology for detecting too small and too large critical infrastructures in satellite images: airports and electrical substations as case study[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021, 14: 12149-12162.
[17] MESVARI M, SHAH-HOSSEINI R. Segmentation of electrical substations using deep convolutional neural network[J]. ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, 2023, 10(4): 495-500.
[18] YANG L X, ZHANG R Y, LI L D, et al. SimAM: a simple, parameter-free attention module for convolutional neural networks[C]∥ International Conference on Machine Learning. ICML: San Diego, 2021: 18-24.
[19] JIANG Y Q, TAN Z Y, WANG J Y, et al. GiraffeDet: a heavy-neck paradigm for object detection[EB/OL]. (2022-02-09)[2025-03-29]. https:∥arxiv. org/abs/2202.04256.
[20] XU X Z, JIANG Y Q, CHEN W H, et al. DAMO-YOLO: a report on real-time object detection design[EB/OL]. (2022-11-23)[2025-03-29]. https:∥arxiv.org/abs/2211.15444.
[21]WANG C Y, MARK LIAO H Y, WU Y H, et al. CSPNet: a new backbone that can enhance learning capability of CNN[C]∥2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). Piscataway: IEEE, 2020: 390-391.
[22] LIU W Z, LU H, FU H T, et al. Learning to upsample by learning to sample[C]∥2023 IEEE/CVF International Conference on Computer Vision (ICCV). Piscataway: IEEE, 2023: 6004-6014.
[23] SHI W Z, CABALLERO J, HUSZÁR F, et al. Real-time single image and video super-resolution using an efficient sub-pixel convolutional neural network[C]∥2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE, 2016: 1874-1883.
[24]WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C]∥ Computer Vision-ECCV 2018. Cham: Springer, 2018: 3-19.
[25] XIA Z F, PAN X R, SONG S J, et al. Vision transformer with deformable attention[C]∥2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE, 2022: 4784-4793.
[26] LIU Z, LIN Y T, CAO Y, et al. Swin transformer: hierarchical vision transformer using shifted windows[C]∥2021 IEEE/CVF International Conference on Computer Vision (ICCV). Piscataway: IEEE, 2021: 999210002.
[27] HE K M, GKIOXARI G, DOLLÁR P, et al. Mask RCNN[C]∥2017 IEEE International Conference on Computer Vision (ICCV). Piscataway: IEEE, 2017: 29802988.
[28] LYU C Q, ZHANG W W, HUANG H A, et al. RTMDet: an empirical study of designing real-time object detectors[EB/OL]. (2022-12-14)[2025-03-29]. https:∥arxiv.org/abs/2212.07784.
[29] ZHANG H, LI F, LIU S L, et al. DINO: DETR with improved denoising anchor boxes for end-to-end object detection[EB/OL]. (2022-07-11)[2025-03-29]. https:∥arxiv.org/abs/2203.03605.

更新日期/Last Update: 2026-01-17