基于部分卷积和动静态融合的无人机辅助绝缘子缺陷检测机制
网络出版日期: 2025-03-19
基金资助
国网新疆电力有限公司电力地下管廊通信覆盖横向项目(SGXJDK00DWJS2310180);国家重大科技专项课题(2022ZD0115802)
版权
UAV auxiliary insulator defect detection mechanism based on partial convolution and dynamic-static fusion
Online published: 2025-03-19
Copyright
在无人机(Unmanned Aerial Vehicle,UAV)自动检测绝缘子图像中的缺陷时,检测算法频繁使用注意力模块会导致模型参数过大,实时性较差。为了实现轻量级、高精度的无人机智能巡检,提出一种基于Vision Transformer(ViT)的无人机绝缘子缺陷检测模型(UAV Insulator Defect Detection Model based on DEtection TRansformer,UID-DETR)。首先,提出的快速重参数模块(Fast Re-parameterization Module,FREP)利用部分卷积(Partial Convolution,PConv)和重参数化卷积(Re-parameterization Convolution,RepConv)减少了冗余计算,高效地提取了空间特征。其次,通过设计高效内尺度交互模块(Efficient Intra-scale Interaction Module,EISI),加强了高级特征的交互。第三,通过静态融合(Static Fusion,STF)和动态融合(Dynamic Fusion,DYF)两种特征融合策略实现了高层级和低层级语义信息的互补结合。大量实验结果验证了所提方法在开源合成雾绝缘子数据集(Synthetic Foggy Insulator Dataset,SFID)和自制绝缘子数据集(Self-made Insulator Dataset,SID)上的有效性。
杨慧婷 , 杨志 , 郭庆瑞 , 李峰 , 郭学让 , 郭治卿 , 汪烈军 . 基于部分卷积和动静态融合的无人机辅助绝缘子缺陷检测机制[J]. 网络空间安全科学学报, 2025 , 3(1) : 86 -99 . DOI: 10.20172/j.issn.2097-3136.250108
When automatically detecting defects in insulator images by unmanned aerial vehicle (UAV) , the frequent use of the attention module by the detection algorithm leads to large model parameters and poor real-time performance. In order to realize lightweight and high-accuracy UAV intelligent inspection, a vision transformer (ViT) based UAV insulator defect detection model based on DEtection TRansformer (UID-DETR) was proposed. Firstly, the proposed fast re-parameterization module (FREP) utilized the partial convolution (PConv) and re-parameterization convolution (RepConv) to reduce redundant computations and extract spatial features efficiently. Secondly, the efficient intra-scale interaction module (EISI) was designed for enhancing the interaction of high-level features. Thirdly, the complementary integration of high-level and low-level semantic information was realized by the feature fusion strategy of static fusion (STF) and dynamic fusion (DYF). Extensive experimental results verified the effectiveness of the proposed method on open-source synthetic foggy insulator dataset (SFID) and self-made insulator dataset (SID).
表 1 目标检测模型在SFID和SID数据集上的性能比较Table 1 Performance comparison of object detection models on SFID and SID datasets |
| Methods | Parameters (M) | FLOPs (G) | SFID | SID | |||||||
| Precision (%) | Recall (%) | mAP50 (%) | Inference time (ms) | Precision (%) | Recall (%) | mAP50 (%) | Inference time (ms) | ||||
| EfficientDet[41] | 11.9 | 47.0 | 99.2 | 92.0 | 96.7 | 148.7 | 96.9 | 86.7 | 93.8 | 112.2 | |
| SSD[42] | 23.7 | 273.6 | 98.4 | 93.9 | 98.7 | 17.0 | 97.1 | 83.5 | 93.1 | 15.0 | |
| CenterNet[43] | 32.7 | 109.7 | 98.6 | 96.8 | 98.3 | 27.5 | 97.4 | 83.0 | 88.5 | 19.2 | |
| Faster R-CNN[44] | 136.7 | 401.7 | 60.0 | 94.9 | 86.3 | 47.4 | 60.7 | 77.7 | 75.4 | 39.7 | |
| YOLOv3-tiny[45] | 12.1 | 18.9 | 96.1 | 96.8 | 98.1 | 2.1 | 97.2 | 79.4 | 89.5 | 2.9 | |
| YOLOv6-s[46] | 16.3 | 44.0 | 98.9 | 97.8 | 99.1 | 3.7 | 91.5 | 84.8 | 89.7 | 6.3 | |
| YOLOv8-m[47] | 25.8 | 78.7 | 99.3 | 99.1 | 99.4 | 5.1 | 96.4 | 86.0 | 93.6 | 7.2 | |
| YOLOv9-m[48] | 32.5 | 130.7 | 98.2 | 98.5 | 99.3 | 11.2 | 96.2 | 88.2 | 94.0 | 24.4 | |
| YOLOv10-m[49] | 16.5 | 63.4 | 97.2 | 97.4 | 99.2 | 6.4 | 96.8 | 83.3 | 90.7 | 7.6 | |
| UID-DETR | 14.1 | 41.4 | 99.4 | 99.2 | 99.4 | 5.7 | 98.3 | 91.8 | 95.4 | 13.5 | |
表 2 消融实验结果Table 2 Results of ablation experiments |
| Baseline | FREP | EISI | STF&DYF | mAP50(%) | F1(%) | Inference time(ms) | Parameters(M) |
| √ | — | — | — | 92.7 | 92.9 | 14.4 | 17.1 |
| √ | √ | — | — | 93.5 | 92.6 | 13.5 | 14.0 |
| √ | √ | √ | — | 93.6 | 93.5 | 14.3 | 14.0 |
| √ | √ | √ | √ | 95.4 | 94.9 | 13.5 | 14.1 |
表 3 与不同轻量骨干网络的比较Table 3 Comparison with different lightweight backbone networks |
| 1 |
SU J, YUAN Y, PRZYSTUPA K, et al. Insulator defect detection algorithm based on improved YOLOv8 for electric power[J]. Signal, Image and Video Processing, 2024, 18 (8-9): 6197- 6209.
|
| 2 |
ZHANG L, LI B, GAO L J. Research on improved YOLOv8 algorithm for insulator defect detection[J]. Journal of Real-Time Image Processing, 2024, 21 (1): 22.
|
| 3 |
LI D, LU Y, GAO Q, et al. LiteYOLO-ID: A lightweight object detection network for insulator defect detection[J]. IEEE Transactions on Instrumentation and Measurement, 2024, 73 (1): 1- 12.
|
| 4 |
QI X,YE S. Defect detection of aerial insulator based on improved lightweight YOLOv5[C]//2023 IEEE 3rd International Conference on Electronic Technology,Communication and Information(ICETCI). IEEE,2023:1833-1839.
|
| 5 |
WU F,MA A J,PAN Y,et al. Receptive field pyramid network for object detection[C]//ICASSP 2020-2020 IEEE International Conference on Acoustics,Speech and Signal Processing(ICASSP). IEEE,2020:1873-1877.
|
| 6 |
MAI J,WANG S,WANG X. Detection of insulator strings in aerial photography based on lightweight CenterNet[C]//2023 8th International Conference on Intelligent Computing and Signal Processing(ICSP). IEEE,2023:821-824.
|
| 7 |
LANHANG C. Defect identification of electricity transmission line insulators based on the improved lightweight network model with computer vision assistance[J]. Heliyon, 2024, 10 (10): e30405.
|
| 8 |
NIU W J,WANG B,BAI B. Detection of insulator defects in transmission lines based on lightweight Convolutional Neural Networks[C]//2024 9th Asia Conference on Power and Electrical Engineering(ACPEE). IEEE,2024:415-419.
|
| 9 |
ZHU X. A lightweight improved YOLOv5 for insulator defect detection[C]//2023 3rd International Conference on Computer Science,Electronic Information Engineering and Intelligent Control Technology(CEI). IEEE,2023:675-678.
|
| 10 |
WANG L,QIN J,YAN K,et al. A lightweight insulator defect detection model based on YOLOv8[C]// 2024 5th International Seminar on Artificial Intelligence,Networking and Information Technology(AINIT). IEEE,2024:2349-2353.
|
| 11 |
LIU J,WANG F. Mixed-attention auto encoder for multi-class industrial anomaly detection[C]//ICASSP 2024-2024 IEEE International Conference on Acoustics,Speech and Signal Processing(ICASSP). IEEE,2024:4120-4124.
|
| 12 |
LIN Y,TIAN L,DU Q,et al. Insulator detection based on improved lightweight rotation detector[C]//2023 4th International Conference on Intelligent Computing and Human-Computer Interaction(ICHCI). IEEE,2023:412-417.
|
| 13 |
BAN M,DING R,ZHANG J,et al. PDD-Net:A precise defect detection network based on point set representation[C]//ICASSP 2022-2022 IEEE International Conference on Acoustics,Speech and Signal Processing(ICASSP). IEEE,2022:2350-2354.
|
| 14 |
LI D, YANG P, ZOU Y. Optimizing insulator defect detection with improved DETR models[J]. Mathematics, 2024, 12 (10): 1507.
|
| 15 |
YUAN W,XIE H,DU S,et al. PAM-DETR:Parallel Attention-MLP for insulator defect detection[C]//2023 International Conference on Sensing,Measurement & Data Analytics in the era of Artificial Intelligence(ICSMD). IEEE,2023:1-6.
|
| 16 |
QU Z, ZANG J, DONG Y, et al. An efficient multi-order cascade distillation model for the detection of small targets and occluded objects in transmission line inspection[J]. Measurement, 2024, 225, 114000.
|
| 17 |
ZHANG Z D, ZHANG B, LAN Z C, et al. FINet: An insulator dataset and detection benchmark based on synthetic fog and improved YOLOv5[J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71, 1- 8.
|
| 18 |
WEI L, JIN J, DENG K, et al. Insulator defect detection in transmission line based on an improved lightweight YOLOv5s algorithm[J]. Electric Power Systems Research, 2024, 233, 110464.
|
| 19 |
WANG Z,LIU Q,ZHOU J. Power insulator defect detection method based on improved YOLOX-s[C]//2024 6th International Conference on Energy,Power and Grid(ICEPG). IEEE,2024:1087-1091.
|
| 20 |
CHEN H,ZHANG Z. Improved YOLOv5 algorithm for power insulator defect detection[C]//2023 13th International Conference on Power,Energy and Electrical Engineering(CPEEE). IEEE,2023:301-305.
|
| 21 |
XIE J,DU Y,YU H,et al. Visible-light insulator defect detection algorithm based on improved YOLOv4-tiny[C]//2023 IEEE International Conference on Power Science and Technology(ICPST). IEEE,2023:986-991.
|
| 22 |
JIANG Z,FU X,CHEN W,et al. Research on insulator defect detection by UAV based on deep learning[C]//2023 5th International Conference on Robotics,Intelligent Control and Artificial Intelligence(RICAI). IEEE,2023:657-661.
|
| 23 |
CHANG R, ZHOU S, ZHANG Y, et al. Research on insulator defect detection based on improved YOLOv7 and multi-UAV cooperative system[J]. Coatings, 2023, 13 (5): 880.
|
| 24 |
TANG S,ZHU R,JI C,et al. Intelligent detection of insulator defects on transmission lines in complex backgrounds based on improved DETR visual perception model[C]//2024 International Conference on Artificial Intelligence and Power Systems(AIPS). IEEE,2024:83-89.
|
| 25 |
TANG C, DONG H, HUANG Y, et al. Foreign object detection for transmission lines based on Swin Transformer V2 and YOLOX[J]. The Visual Computer, 2024, 40 (5): 3003- 3021.
|
| 26 |
LI Y, FENG D, ZHANG Q, et al. HRD-YOLOX based insulator identification and defect detection method for transmission lines[J]. IEEE Access, 2024, 12, 22649- 22661.
|
| 27 |
HE M, QIN L, DENG X, et al. MFI-YOLO: Multi-fault insulator detection based on an improved YOLOv8[J]. IEEE Transactions on Power Delivery, 2024, 39 (1): 168- 179.
|
| 28 |
ZOU X,ZHOU Y. DeFCN-nano:An end-to-end real-time object detection for insulator defects[C]//2023 IEEE International Conference on Robotics and Biomimetics(ROBIO). IEEE,2023:1-6.
|
| 29 |
ZHANG Z,JIANG Q,CHU X. Self-exploding insulator detection for transmission lines based on improved YOLOv5s algorithm[C]//2024 2nd International Conference on Signal Processing and Intelligent Computing(SPIC). IEEE,2024:52-55.
|
| 30 |
LIU X,RAO Z,ZHANG Y,et al. UAVs images based real-time insulator defect detection with Transformer deep learning[C]//2023 IEEE International Conference on Robotics and Biomimetics(ROBIO). IEEE,2023:1-6.
|
| 31 |
CHEN J,KAO S,HE H,et al. Run,don't walk:Chasing higher FLOPS for faster neural networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE,2023:12021-12031.
|
| 32 |
DING X,ZHANG X,MA N,et al. RepVGG:Making VGG-style ConvNets great again[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE,2021:13733-13742.
|
| 33 |
ZHAO Y,LV W,XU S,et al. DETRs beat YOLOs on real-time object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE,2024:16965-16974.
|
| 34 |
SHAKER A,MAAZ M,RASHEED H,et al. SwiftFormer:Efficient additive attention for Transformer-based real-time mobile vision applications[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. IEEE,2023:17425-17436.
|
| 35 |
LIU W,LU H,FU H,et al. Learning to upsample by learning to sample[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. IEEE,2023:6027-6037.
|
| 36 |
WANG J,CHEN K,XU R,et al. CARAFE:Content-aware reassembly of features[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. IEEE,2019:3007-3016.
|
| 37 |
LU H,LIU W,FU H,et al. FADE:Fusing the assets of decoder and encoder for task-agnostic upsampling[C]//European Conference on Computer Vision. Cham:Springer Nature Switzerland,2022:231-247.
|
| 38 |
LU H, LIU W, YE Z, et al. SAPA: Similarity-aware point affiliation for feature upsampling[J]. Advances in Neural Information Processing Systems, 2022, 35, 20889- 20901.
|
| 39 |
TAO X, ZHANG D, WANG Z, et al. Detection of power line insulator defects using aerial images analyzed with convolutional neural networks[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2018, 50 (4): 1486- 1498.
|
| 40 |
ANDREL V S,CHAVES T,FELIX H,et al. Unifying public datasets for insulator detection and fault classification in electrical power lines[DB/OL](2020)[2025-3-10]. https://github.com/heitorcfelix/public-insulator-datasets.
|
| 41 |
TAN M,PANG R,LE Q V. EfficientDet:Scalable and efficient object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE,2020:10781-10790.
|
| 42 |
LIU W,ANGUELOV D,ERHAN D,et al. SSD:Single shot multibox detector[C]//Computer Vision–ECCV 2016:14th European Conference. Cham:Springer International Publishing,2016:21-37.
|
| 43 |
ZHOU X,KOLTUN V,KRÄHENBÜHL P. Tracking objects as points[C]//European Conference on Computer Vision. Cham:Springer International Publishing,2020:474-490.
|
| 44 |
REN S, HE K, 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.
|
| 45 |
FARHADI A,REDMON J. YOLOv3:An incremental improvement[C]//Computer Vision and Pattern Recognition. Berlin,Heidelberg:Springer,2018,1804:1-6.
|
| 46 |
LI C, LI L, JIANG H, et al. YOLOv6: A single-stage object detection framework for industrial applications[J]. ArXiv preprint, ArXiv:, 2209, 02976, 2022.
|
| 47 |
VARGHESE R,SAMBATH M. YOLOv8:A novel object detection algorithm with enhanced performance and robustness[C]//2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS),2024:1-6.
|
| 48 |
WANG C Y, YEH I H, LIAO H Y M. YOLOv9: Learning what you want to learn using programmable gradient information[J]. ArXiv preprint, ArXiv:, 2402, 13616, 2024.
|
| 49 |
WANG A, CHEN H, LIU L, et al. YOLOv10: Real-time end-to-end object detection[J]. ArXiv preprint, ArXiv:, 2405, 14458, 2024.
|
| 50 |
DONG X,BAO J,CHEN D,et al. CSWin Transformer:A general vision transformer backbone with cross-shaped windows[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE,2022:12124-12134.
|
| 51 |
LIU Z,LIN Y,CAO Y,et al. Swin Transformer:Hierarchical Vision Transformer using shifted windows[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. IEEE,2021:10012-10022.
|
| 52 |
LIU X,PENG H,ZHENG N,et al. EfficientViT:Memory efficient Vision Transformer with Cascaded Group Attention[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE,2023:14420-14430.
|
/
| 〈 |
|
〉 |