Academic Journal of Computing & Information Science, 2026, 9(3); doi: 10.25236/AJCIS.2026.090303.
Peiqi He1, Jinhao Yuan1
1School of Mechanical Engineering, University of Shanghai for Science and Technology, Shanghai, China
To address issues in outdoor power facility detection such as strong environmental interference, large variations in target scale, and slender structures being easily obscured by the background, this paper reconstructs a power facility detection dataset and proposes an improved framework for outdoor power facility detection based on YOLOv11n. The proposed method systematically redesigns the network architecture and feature modeling approach in terms of multi-scale feature modeling, orientation-aware modeling for slender targets, and robust feature modeling under complex backgrounds. In particular, the feature modeling strategy for handling complex background interference is further enhanced. Experimental results demonstrate that the proposed method significantly outperforms the baseline model on the self-constructed power facility dataset, with [email protected] and [email protected]:0.95 increased by 4.3% and 10.3%, respectively. Meanwhile, it achieves comparable performance to YOLOv11n on several public datasets, demonstrating strong generalization capability and providing solid evidence that the proposed architecture can more effectively accomplish intelligent detection of outdoor power facilities.
Power facility detection; YOLOv11n; Multi-scale features; Attention mechanism; Extreme outdoor environments
Peiqi He, Jinhao Yuan. Attention-Enhanced Multi-Scale Feature Modeling for Slender Power Facility Detection in Complex Outdoor Environments. Academic Journal of Computing & Information Science (2026), Vol. 9, Issue 3: 17-29. https://doi.org/10.25236/AJCIS.2026.090303.
[1] Korolev I , Zakrevsky A , Vasileva N V .Protection of power facilities personnel from electric fields of industrial frequency[J].2023 5th International Youth Conference on Radio Electronics, Electrical and Power Engineering (REEPE), 2023, 5:1-5.DOI:10.1109/REEPE57272.2023.10086848.
[2] CHEN Jiaqi, SHENG Shuang, LIN Jiaxi, et al. Zero sample power grid equipment ontology defect grade idetification method based on knowledge enhanced large language model[J].Chinese High Technology Letters,2025,35(04):429-439.doi:10. 3772/ j. issn. 1002-0470. 2025. 04. 010.
[3] WANG H, MENG F. Research on power equipment recognition method based on image processing[J/OL]. EURASIP Journal on Image and Video Processing, 2019.
[4] BAI K, ZHOU Y, CUI Z, et al. HOG-SVM-Based Image Feature Classification Method for Sound Recognition of Power Equipments[J/OL]. Energies, 2022, 15(12): 4449.
[5] LEI L, WU J, ZHENG S, et al. Recognition of Power Equipment Based on Multitask Sparse Representation[J/OL]. Scientific Programming, 2021, 2021: 1-7.
[6] ARASTOUNIA M, LICHTI D. Automatic Object Extraction from Electrical Substation Point Clouds [J/OL]. Remote Sensing, 2015: 15605-15629.
[7] ENGELCKE M, RAO D, WANG D Z, et al. Vote3Deep: Fast object detection in 3D point clouds using efficient convolutional neural networks[C/OL]//2017 IEEE International Conference on Robotics and Automation (ICRA), Singapore, Singapore. 2017.
[8] YU W, XI J, WU Z, et al. A METHOD FOR EXTRACTING SUBSTATION EQUIPMENT BASED ON UAV LASER SCANNING POINT CLOUDS[J/OL]. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2020, XLIV-4/W3-2020: 413-419.
[9] ZHANG R, YANG B, XIAO W, et al. Automatic Extraction of High-Voltage Power Transmission Objects from UAV Lidar Point Clouds[J/OL]. Remote Sensing, 2019: 2600.
[10] ARASTOUNIA M, LICHTI D D. Segmentation of planar surfaces in LiDAR point clouds of an electrical substation by exploring the structure of points neighbourhood[J/OL]. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2014, XL-5: 55-62.
[11] YUAN Q, LUO Y, WANG H. 3D point cloud recognition of substation equipment based on plane detection[J/OL]. Results in Engineering, 2022: 100545.
[12] GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation[C/OL]//2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA. 2014.
[13] GIRSHICK R. Fast R-CNN[C/OL]//2015 IEEE International Conference on Computer Vision (ICCV), Santiago, Chile. 2015.
[14] REDMON J, DIVVALA S, GIRSHICK R, et al. You Only Look Once: Unified, Real-Time Object Detection[C/OL]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA. 2016.
[15] REDMON J, FARHADI A. YOLO9000: Better, Faster, Stronger[C/OL]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI. 2017.
[16] LIU W, ANGUELOV D, ERHAN D, et al. SSD: Single Shot MultiBox Detector[M/OL]//Computer Vision – ECCV 2016,Lecture Notes in Computer Science. 2016: 21-37.
[17] ZHOU F, MA Y, WANG B, et al. Dual-channel convolutional neural network for power edge image recognition[J/OL]. Journal of Cloud Computing, 2021, 10(1).
[18] XIONG X, XU S, WU W, et al. Identification of Electrical Equipment Based on Faster LSTM-CNN Network[C/OL]//2020 IEEE International Conference on Networking, Sensing and Control (ICNSC), Nanjing, China. 2020:1-6.
[19] WANG H, YANG G, LI E, et al. High-Voltage Power Transmission Tower Detection Based on Faster R-CNN and YOLO-V3[C/OL]//2019 Chinese Control Conference (CCC), Guangzhou, China. 2019.
[20] Wu Y , Xiao F , Liu F ,et al.A Visual Fault Detection Algorithm of Substation Equipment Based on Improved YOLOv5[J].Applied Sciences (2076-3417), 2023, 13(21).
[21] Wang Q , Yang L , Zhou B ,et al.YOLO-SS-Large: A Lightweight and High-Performance Model for Defect Detection in Substations[J].Sensors (14248220), 2023, 23(19).
[22] Yang Y , Yang S , Li C ,et al.Insulator defect detection under extreme weather based on synthetic weather algorithm and improved YOLOv7[J].High Voltage, 2025, 10(1)
[23] Shao, H., Zeng, Q., Hou, Q. et al. MCANet: Medical Image Segmentation with Multi-scale Cross-axis Attention. Mach. Intell. Res. 22, 437–451 (2025).
[24] AZAD R, NIGGEMEIER L, HUTTEMANN M, et al. Beyond Self-Attention: Deformable Large Kernel Attention for Medical Image Segmentation[J]. Computer Vision and Pattern Recognition,2023.
[25] Dai W , Liu R , Wu Z ,et al.Exploiting Scale-Variant Attention for Segmenting Small Medical Objects[J]. Image and Video Processing,2024.
[26] TONG Z, CHEN Y, XU Z, et al. Wise-IoU: Bounding Box Regression Loss with Dynamic Focusing Mechanism[J]. Computer Vision and Pattern Recognition,2023.
[27] HU J, SHEN L, ALBANIE S, et al. Squeeze-and-Excitation Networks[J/OL]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020: 2011-2023.