Infrared and Laser Engineering, Volume. 50, Issue 5, 20200309(2021)

Application of complex bidimensional empirical mode decomposition in SAR target recognition

Zhichao Liu1,2 and Baida Qu1、*
Author Affiliations
  • 1School of Internet of Things Engineering, Jiangnan University, Wuxi 214122, China
  • 2School of Internet of Things Engineering, Wuxi Taihu University, Wuxi 214064, China
  • show less
    Figures & Tables(11)
    Illustration of decomposition of SAR images via BEMD
    SAR target recognition method based on feature extraction by C-BEMD
    Illustration of MSTAR targets
    Recognition results of each class by the proposed method under SOC
    Results under pitch angle variances
    Results under noise corruption
    • Table 1. Description of SOC

      View table
      View in Article

      Table 1. Description of SOC

      ClassTraining set (17°)Test sample set (15°)
      Configuration (SN)Scale of samplesConfiguration (SN)Scale of samples
      BMP295632339563195
      BTR70233196
      T72132232132196
      812195
      S7191
      T62299273
      BRDM2298274
      BTR60256195
      ZSU23/4299274
      D7299274
      ZIL131299274
      2S1299274
    • Table 2. Description of configruation variance

      View table
      View in Article

      Table 2. Description of configruation variance

      ClassTraining set (17°)Test sample set (15°)
      Configuration (SN)Scale of samplesConfiguration (SN)Scale of samples
      BMP295632339566196
      C21196
      BTR70233196
      T72132232812195
      S7191
    • Table 3. Description of pitch angle variance

      View table
      View in Article

      Table 3. Description of pitch angle variance

      ClassTraining setTest sample set
      Pitch angle/(°)Scale of samplesPitch angle/(°)Scale of samples
      2S11729930288
      45303
      BDRM21729830287
      45303
      ZSU23/41729930288
      45303
    • Table 4. Results under SOC

      View table
      View in Article

      Table 4. Results under SOC

      Method typeAverage recognition rate
      Proposed method99.34%
      NMF98.26%
      Mono98.74%
      BEMD99.04%
      CNN99.08%
    • Table 5. Results under configuration variance

      View table
      View in Article

      Table 5. Results under configuration variance

      Method typeAverage recognition rate
      Proposed method99.12%
      NMF97.26%
      Mono96.36%
      BEMD98.27%
      CNN98.68%
    Tools

    Get Citation

    Copy Citation Text

    Zhichao Liu, Baida Qu. Application of complex bidimensional empirical mode decomposition in SAR target recognition[J]. Infrared and Laser Engineering, 2021, 50(5): 20200309

    Download Citation

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

    Category: Image processing

    Received: Dec. 2, 2020

    Accepted: --

    Published Online: Aug. 13, 2021

    The Author Email: Qu Baida (qbd518@aliyun.com)

    DOI:10.3788/IRLA20200309

    Topics