Journals Articles Conferences News About CLP
Submit a paper Email Alert
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
2D MaterialsTransformation opticsQuantum PhotonicsSmall lasersSemiconductor UV PhotonicsSupersymmetric microring laser arrays

Chinese Journal of Lasers, Volume. 47, Issue 7, 710002(2020)

Research on SAR Sea Significant Wave Height Inversion Method Based on ELM Model

Wang Xiaochen1,2, He Dongxu1、*, and Liu Bingxuan2,3
Author Affiliations
  • 1Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100101, China
  • 2Zhejiang Key Laboratory,Deqing Academy of Satellite Application, Huzhou 313200, Zhejiang, China
  • 3Beijing Branch of Chinese Academy of Sciences, Beijing 100101, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (15)
    Equations (0)
    References (21)
    Cited By (1)
    Tools
    Paper Information
    Topics
    Figures & Tables(15)
    Global sea surface significant wave height distribution of January, 2011

    Fig. 1. Global sea surface significant wave height distribution of January, 2011

    Download full size
    Buoys location used in this paper

    Fig. 2. Buoys location used in this paper

    Download full size
    SAR imagery of wave stripes. (a) Clear wave stripes; (b) fuzzy wave stripe

    Fig. 3. SAR imagery of wave stripes. (a) Clear wave stripes; (b) fuzzy wave stripe

    Download full size
    Comparison results of ECMWF reanalysis data and buoy data

    Fig. 4. Comparison results of ECMWF reanalysis data and buoy data

    Download full size
    Statistical histogram of significant wave height for model training and verification in the case of mass-matched dataset

    Fig. 5. Statistical histogram of significant wave height for model training and verification in the case of mass-matched dataset

    Download full size
    Validation of significant wave height retrieval algorithm based on ELM in the case of mass-matched dataset

    Fig. 6. Validation of significant wave height retrieval algorithm based on ELM in the case of mass-matched dataset

    Download full size
    Validation of significant wave height retrieval algorithm based on CWAVE in the case of mass-matched dataset

    Fig. 7. Validation of significant wave height retrieval algorithm based on CWAVE in the case of mass-matched dataset

    Download full size
    Statistical histogram of significant wave height for model training and verification in the case of less-matched dataset

    Fig. 8. Statistical histogram of significant wave height for model training and verification in the case of less-matched dataset

    Download full size
    Validation of significant wave height retrieval algorithm based on ELM in the case of less-matched dataset

    Fig. 9. Validation of significant wave height retrieval algorithm based on ELM in the case of less-matched dataset

    Download full size
    Validation of significant wave height retrieval algorithm based in CWAVE in the case of less-matched dataset

    Fig. 10. Validation of significant wave height retrieval algorithm based in CWAVE in the case of less-matched dataset

    Download full size
    Retrieval error variation with match-up samples

    Fig. 11. Retrieval error variation with match-up samples

    Download full size
    ELM model accuracy

    Fig. 12. ELM model accuracy

    Download full size

    • Table 1. Parameters of ENVISAT ASAR wave mode

      View table

      Table 1. Parameters of ENVISAT ASAR wave mode

      ParameterValue
      Band /GHz5。4
      PolarizationHH or VV
      Incidence angle /(°)14。1 to 42.3
      Azimuth resolution /m4
      Range resolution /m7
      Swath /km5

    • Table 2. Comparison of our proposed algorithm and CWAVE algorithm in the case of mass-matched dataset

      View table

      Table 2. Comparison of our proposed algorithm and CWAVE algorithm in the case of mass-matched dataset

      Retrieval resultsAlgorithm inthis paperCWAVEalgorithm
      Correlation0.870.91
      Bias /m0.390
      RetrievalaccuracyRMSE /m0.390.45
      SI00.01
      Training efficiency(running time) /s0.0220.514

    • Table 3. Comparison of our proposed algorithm and CWAVE algorithm in the case of less-matched dataset

      View table

      Table 3. Comparison of our proposed algorithm and CWAVE algorithm in the case of less-matched dataset

      Retrieval resultsAlgorithm inthis paperCWAVEalgorithm
      Correlation0.59-0.38
      Bias /m0.18-0.63
      RetrievalaccuracyRMSE /m0.181.5
      SI00.75
      Training efficiency(running time) /s0.0080.318
    Tools

    Get Citation

    Copy Citation Text

    Wang Xiaochen, He Dongxu, Liu Bingxuan. Research on SAR Sea Significant Wave Height Inversion Method Based on ELM Model[J]. Chinese Journal of Lasers, 2020, 47(7): 710002

    Download Citation

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

    Category: remote sensing and sensor

    Received: Jan. 3, 2020

    Accepted: --

    Published Online: Jul. 10, 2020

    The Author Email: Dongxu He (hedx@aircas.ac.cn)

    DOI:10.3788/CJL202047.0710002

    Topics

    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology