计算机科学 ›› 2024, Vol. 51 ›› Issue (11A): 240100003-6.doi: 10.11896/jsjkx.240100003
宋尚泽1, 李莉1, 田野2, 白洁2
SONG Shangze1, LI Li1, TIAN Ye2, BAI Jie2
摘要： 在电力线路检测领域,准确检测细微裂纹和微小破损等微小损伤至关重要。这些轻微损坏往往因其规模小和背景复杂性而被忽视,如果不及时识别和解决,可能会升级为重大安全隐患。为了应对这一挑战,本研究设计了 PowerScreen-YOLOv8(PS-YOLOv8) 模型。该模型与原始YOLOv8相比,对电力巡检中的小目标检测有了很大进步,通过集成了6项关键改进,以提高复杂环境中的检测精度。该研究通过严格的测试和针对领先算法的基准测试证明了该模型的优越性。PS-YOLOv8 获得了90.3%的准确率并且在现实无人机捕获场景中具有经过验证的稳健性,代表了电力线路检测技术的重大飞跃,为基础设施维护提供了更可靠、更高效、更安全的方法。
中图分类号:
| [1]LIU Z Y,ZHAO X D,QI H C,et al.Prospects of UAV Power Inspection Technology in the New Era [J].Southern Energy Construction,2019,6(4):1-5. [2]TONG W G,YUAN J S,LI B S.A Review on the Application of Image Processing Technology in Helicopter Inspection of Transmission Lines [J].Power Grid Technology,2010,34(12):204-208. [3]JIANG P,ERGU D,LIU F,et al.A Review of Yolo algorithm developments[J].Procedia Computer Science,2022,199:1066-1073. [4]ABOAH A,WANG B,BAGCI U,et al.Real-time multi-class helmet violation detection using few-shot data sampling technique and yolov8[C]//Proceedings of the IEEE/CVF Confe-rence on Computer Vision and Pattern Recognition.2023:5349-5357. [5]YUAN Y F,ZHOU Z C,ZHANG C,et al.Research on Lightweight Object Detection and Fault Recognition Methods for Power Inspection [J].Electric Power Information and Communication Technology,2022,20(8):29-37. [6]XUE K T,SAVKINE J,GAO J L.Power Inspection Image Super-Resolution Reconstruction Algorithm Based on VDRCNN [J].Journal of Jilin University (Information Science Edition),2023,41(3):530-538. [7]HU N,HUANG Z H.Power Tower Key Point Detection Based on Deep Learning [J].Science and Innovation,2022(7):69-74. [8]HUANG W W,YUAN B,WANG B,et al.Unattended Power Inspection System Based on Image Fusion and H.265 [J].Mo-dern Electronic Technology,2022,45(6):131-136. [9]ZHANG W H,ZHANG Z Z,XIE F g,et al.Research on Fault Recognition Algorithm for Photovoltaic Panels Based on SSD and Infrared Video [J].Computer Knowledge and Technology,2022,18(35):90-92. [10]LIU Y,SUN P,WERGELES N,et al.A survey and performance evaluation of deep learning methods for small object detection[J].Expert Systems with Applications,2021,172:114602. [11]QI Y,HE Y,QI X,et al.Dynamic Snake Convolution based on Topological Geometric Constraints for Tubular Structure Segmentation[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2023:6070-6079. [12]HAN K,XIAO A,WU E,et al.Transformer in transformer[J].Advances in Neural Information Processing Systems,2021,34:15908-15919. [13]ZHU L,WANG X,KE Z,et al.BiFormer:Vision Transformer with Bi-Level Routing Attention[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2023:10323-10333. [14]MISRA D.Mish:A self regularized non-monotonic activationfunction[J].arXiv:1908.08681,2019. [15]JOCHER G,STOKEN A,BOROVEC J,et al.ultralytics/yolov5:v4.0-nn.SiLU () activations,Weights & Biases logging,PyTorch Hub integration [EB/OL].https://github.com/ultralytics/yolov5. [16]LI H,LI J,WEI H,et al.Slim-neck by GSConv:A better design paradigm of detector architectures for autonomous vehicles[J].arXiv:2206.02424,2022. [17]TONG Z,CHEN Y,XU Z,et al.Wise-IoU:Bounding Box Regression Loss with Dynamic Focusing Mechanism[J].arXiv:2301.10051,2023. [18]CHEN X,LIANG C,HUANG D,et al.Symbolic discovery of optimization algorithms[J].arXiv:2302.06675,2023. [19]FAN Y L.Research on Deep Learning-based Foreign Object Detection Methods in Power Inspection Images [D].Hubei University of Technology,2020. [20]LI B,QU L Y,ZHU X S,et al.Insulator Defect Detection Based on Multi-scale Feature Fusion [J].Transactions of Electrical Engineering,2023,38(1):60-70. [21]KANG T A,WANG B L,LIU S H,et al.A Review on Deep Learning Detection Methods for Transmission Line Fittings and Their Defects [J].Electric Power Information and Communication Technology,2022,20(11):1-12. [22]DING N N,HU X X,WU Y C,et al.Research on Deep Learning-based Detection Methods for External Force Damage to Transmission Lines [J].Software Engineering,2022,25(1):14-17. [23]GU X D,TANG D H,HUANG X H.Defect Detection and Recognition in Power Grid Inspection Images Based on Deep Learning [J].Power System Protection and Control,2021,49(5):7. [24]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.2020:10781-10790. [25]BENJUMEA A,TEETI I,CUZZOLIN F,et al.YOLO-Z:Improving small object detection in YOLOv5 for autonomous vehicles[J].arXiv:2112.11798,2021. [26]WANG G,LIU Z,SUN H,et al.Yolox-BTFPN:An anchor-free conveyor belt damage detector with a biased feature extraction network[J].Measurement,2022,200:111675. [27]ZHU X,SU W,LU L,et al.Deformable DETR:DeformableTransformers for End-to-End Object Detection[C]//International Conference on Learning Representations.2021. [28]YE G,QU J,TAO J,et al.Autonomous surface crack identification of concrete structures based on the YOLOv7 algorithm[J].Journal of Building Engineering,2023,73:106688. [29]YU H,YUN L,CHEN Z,et al.A Small Object Detection Algorithm Based on Modulated Deformable Convolution and Large Kernel Convolution[J/OL].Computational Intelligence and Neuroscience,2023.https://pubmed.ncbi.nlm.nih.gov/36733786/. [30]LI K,WANG Y,HU Z.Improved YOLOv7 for Small Object Detection Algorithm Based on Attention and Dynamic Convolution[J].Applied Sciences,2023,13(16):9316. |
|
||
首页