Optics and Precision Engineering, Volume. 33, Issue 9, 1446(2025)

Knowledge graph-based method for infrared target component recognition

Haiyi LIU, Zhengzhou LI*, Aoran LI, and Haitao LIU
Author Affiliations
  • School of Microelectronics and Communication Engineering, Chongqing University, Chongqing401331, China
  • show less
    Figures & Tables(13)
    Implementation process of whole to component prediction(WCP) method
    Structure of acknowledge graph-based component detection algorithm
    Structure of CAM
    Example of aircraft attitude
    Flow chart of knowledge graph inference
    Label format
    Equipment and experimental environment for infrared target recognition
    Example of each attitude and distance of target
    Performance comparison of proposed algorithm with original algorithm
    mAP curves for each component
    Comparison of aircraft component recognition results at various attitudes and distances
    • Table 1. Results of ablation experiment

      View table
      View in Article

      Table 1. Results of ablation experiment

      BaseWCPCAMSLDPRmAP
      82.487.487.3
      86.493.591.8
      88.194.091.0
      81.890.287.5
      93.590.692.2
    • Table 2. Comparison results of different algorithms

      View table
      View in Article

      Table 2. Comparison results of different algorithms

      类别YOLOv8tinyHTCProposed
      飞机头0.9440.9450.9520.907
      驾驶舱0.8860.8690.8870.919
      左机翼0.9330.9530.9460.955
      右机翼0.8840.9070.9000.969
      左尾翼0.8920.8910.9320.896
      右尾翼0.7950.8240.7860.905
      左垂翼0.9820.9650.9800.927
      右垂翼0.9480.9550.9330.904
      发动机0.5980.6820.6940.917
      mAP0.8730.8960.8900.922
    Tools

    Get Citation

    Copy Citation Text

    Haiyi LIU, Zhengzhou LI, Aoran LI, Haitao LIU. Knowledge graph-based method for infrared target component recognition[J]. Optics and Precision Engineering, 2025, 33(9): 1446

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category:

    Received: Feb. 13, 2025

    Accepted: --

    Published Online: Jul. 22, 2025

    The Author Email: Zhengzhou LI (lizhengzhou@cqu.edu.cn)

    DOI:10.37188/OPE.20253309.1446

    Topics