• Laser & Optoelectronics Progress
  • Vol. 59, Issue 2, 0210005 (2022)
Jianchen Gao, Jiahong Zhang*, Yingna Li, and Chuan Li
Author Affiliations
  • Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming , Yunnan 650000, China
  • show less
    DOI: 10.3788/LOP202259.0210005 Cite this Article Set citation alerts
    Jianchen Gao, Jiahong Zhang, Yingna Li, Chuan Li. Insulator Burst Fault Identification Based on YOLOv4[J]. Laser & Optoelectronics Progress, 2022, 59(2): 0210005 Copy Citation Text show less
    YOLOv4 network structure
    Fig. 1. YOLOv4 network structure
    PANet network structure. (a) FPN backbone; (b) bottom-up path augmentation
    Fig. 2. PANet network structure. (a) FPN backbone; (b) bottom-up path augmentation
    CIOU schematic diagram
    Fig. 3. CIOU schematic diagram
    Data augmentation
    Fig. 4. Data augmentation
    Marking of insulator samples
    Fig. 5. Marking of insulator samples
    Mosaic data augmentation
    Fig. 6. Mosaic data augmentation
    Convergence curve of loss function
    Fig. 7. Convergence curve of loss function
    Simulation test results. (a) No background interference; (b)(c) with towers interference; (d) single target; (e) double target; (f) multi-target
    Fig. 8. Simulation test results. (a) No background interference; (b)(c) with towers interference; (d) single target; (e) double target; (f) multi-target
    P-R curves of K-means++ YOLOv4 model. (a) Insulator identification; (b) insulator burst fault identification
    Fig. 9. P-R curves of K-means++ YOLOv4 model. (a) Insulator identification; (b) insulator burst fault identification
    Comparison of performance indicators of different detection algorithms
    Fig. 10. Comparison of performance indicators of different detection algorithms
    TargetTPFPFN
    Insulator4490709228
    Insulator damage2654713
    Table 1. Test results
    Jianchen Gao, Jiahong Zhang, Yingna Li, Chuan Li. Insulator Burst Fault Identification Based on YOLOv4[J]. Laser & Optoelectronics Progress, 2022, 59(2): 0210005
    Download Citation