• Laser & Optoelectronics Progress
  • Vol. 58, Issue 4, 0407001 (2021)
Kaibin Feng1、2, Rufeng Tang1, Rongwang Li1、3、*, and Yuqiang Li1、3
Author Affiliations
  • 1Yunnan Observatories, Chinese Academy of Sciences, Kunming, Yunnan 650216, China
  • 2University of Chinese Academy of Sciences, Beijing 100049, China
  • 3Key Laboratory of Space Object & Debris Observation, Chinese Academy of Sciences, Nanjing, Jiangsu 210034, China;
  • show less
    DOI: 10.3788/LOP202158.0407001 Cite this Article Set citation alerts
    Kaibin Feng, Rufeng Tang, Rongwang Li, Yuqiang Li. Application of Deep Learning in Data Processing of Satellite Laser Ranging[J]. Laser & Optoelectronics Progress, 2021, 58(4): 0407001 Copy Citation Text show less
    Flow chart of laser ranging data processing based on deep learning
    Fig. 1. Flow chart of laser ranging data processing based on deep learning
    Collection date of experimental data and target quantity
    Fig. 2. Collection date of experimental data and target quantity
    Results of Ajisai laser ranging satellite. (a) Original data; (b)result after treatment by proposed algorithm; (c) zoomed result
    Fig. 3. Results of Ajisai laser ranging satellite. (a) Original data; (b)result after treatment by proposed algorithm; (c) zoomed result
    Results of space debris 001575. (a) Original data; (b)result after treatment by proposed algorithm; (c) zoomed result
    Fig. 4. Results of space debris 001575. (a) Original data; (b)result after treatment by proposed algorithm; (c) zoomed result
    Results of space debris 20580. (a) Original data; (b)result after treatment by proposed algorithm; (c) zoomed result
    Fig. 5. Results of space debris 20580. (a) Original data; (b)result after treatment by proposed algorithm; (c) zoomed result
    Cumulative distribution of coincidence percentage
    Fig. 6. Cumulative distribution of coincidence percentage
    Number of imagesTraining accuracy /%Test accuracy /%F1-score
    2000.99990.82960.7731
    5000.99990.84710.8385
    7000.99990.89170.8649
    10000.99990.88540.8636
    15000.99990.89330.8651
    17740.99940.89330.8615
    Table 1. Performance of deep neural networks under different numbers of ranging residual images
    Target nameNC numberGraz numberCross numberCross rate /%
    Ajisai45292452014507399.72
    0015753177100
    2058021612310081.30
    Table 2. Recognization result comparison of proposed algorithm and Graz method
    Image numberNC numberGraz numberCross numberCross rate /%
    47243685293427289.80
    Table 3. Comparison of batch ranging results
    Kaibin Feng, Rufeng Tang, Rongwang Li, Yuqiang Li. Application of Deep Learning in Data Processing of Satellite Laser Ranging[J]. Laser & Optoelectronics Progress, 2021, 58(4): 0407001
    Download Citation