• Laser & Optoelectronics Progress
  • Vol. 62, Issue 5, 0512006 (2025)
Yulong Suo1,2,*, Haifeng Zhang1,3, Bao Zhang4, Jiefeng Sun1,2..., Si Qin1 and Mingliang Long1|Show fewer author(s)
Author Affiliations
  • 1Shanghai Astronomical Observatory, Chinese Academy of Sciences, Shanghai 200030, China
  • 2School of Astronomy and Space Science, University of Chinese Academy of Sciences, Beijing 100049, China
  • 3Key Laboratory of Space Object and Debris Observation, Chinese Academy of Sciences, Nanjing 210008, Jiangsu , China
  • 4People's Liberation Army 61711, Kashi 844000, Xinjiang , China
  • show less
    DOI: 10.3788/LOP241257 Cite this Article Set citation alerts
    Yulong Suo, Haifeng Zhang, Bao Zhang, Jiefeng Sun, Si Qin, Mingliang Long. Application of Deep Learning Algorithm for Automatic Identification of Key Points for Laser Beam Pointing[J]. Laser & Optoelectronics Progress, 2025, 62(5): 0512006 Copy Citation Text show less
    Two typical kinds of laser beam images. (a) Laser beam image in clear weather; (b) laser beam image when laser beam hitting thin cloud
    Fig. 1. Two typical kinds of laser beam images. (a) Laser beam image in clear weather; (b) laser beam image when laser beam hitting thin cloud
    Distribution of key points in laser beam dataset
    Fig. 2. Distribution of key points in laser beam dataset
    Partial images of dataset after data augmentation
    Fig. 3. Partial images of dataset after data augmentation
    Structure of YOLOv8-Pose model
    Fig. 4. Structure of YOLOv8-Pose model
    Comparison of key point recognition performance using different methods. (a) Traditional image processing method; (b) YOLOv8-Pose
    Fig. 5. Comparison of key point recognition performance using different methods. (a) Traditional image processing method; (b) YOLOv8-Pose
    Identification results of laser beam key point at different conditions by proposed algorithm. (a) Identification result with star interference; (b) identification result when laser beam hitting thin cloud
    Fig. 6. Identification results of laser beam key point at different conditions by proposed algorithm. (a) Identification result with star interference; (b) identification result when laser beam hitting thin cloud
    Identification results of laser beam key point while Galileo207 satellite ranging. (a) Identification result of laser beam key point while real-time ranging; (b) real-time ranging result of Galileo207 satellite
    Fig. 7. Identification results of laser beam key point while Galileo207 satellite ranging. (a) Identification result of laser beam key point while real-time ranging; (b) real-time ranging result of Galileo207 satellite
    InterferenceAtmosphericTotalDeviation below 2 pixelDeviation range is 2,3 pixel
    StarGood transparency99990
    Poor transparency41374
    Thin cloud36351
    No starGood transparency1761742
    Poor transparency95896
    Thin cloud48462
    Table 1. Test results of validation set
    Yulong Suo, Haifeng Zhang, Bao Zhang, Jiefeng Sun, Si Qin, Mingliang Long. Application of Deep Learning Algorithm for Automatic Identification of Key Points for Laser Beam Pointing[J]. Laser & Optoelectronics Progress, 2025, 62(5): 0512006
    Download Citation