Laser & Optoelectronics Progress, Volume. 57, Issue 22, 221020(2020)

Multi-Resolution Selection of SAR Images and Target Recognition Based on Nonlinear Correlation Information Entropy

Jie He* and Xin Chen
Author Affiliations
  • College of Mobile Telecommunications, Chongqing University of Post and Telecommunications, Chongqing 401520, China
  • show less
    Figures & Tables(11)
    Diagram of multi-resolution representations. (a) 0.3m; (b) 0.4m; (c) 0.5m; (d) 0.6m
    Selection of multi-resolution representations and target recognition method based on nonlinear correlation information entropy for SAR images
    Procedure of data reduction based on PCA
    Optical image and SAR image of MSTAR targets
    Classification result of 10-class targets
    Average recognition rates under noise corruption interference
    • Table 1. NCIE value under different subsets of resolution

      View table

      Table 1. NCIE value under different subsets of resolution

      Subset of resolutionNCIE
      (0.3, 0.35, 0.45, 0.5, 0.55)0.88
      (0.3, 0.35, 0.4, 0.5)0.82
      (0.3, 0.35, 0.45, 0.55)0.79
      (0.35, 0.4, 0.45)0.71
      (0.3, 0.35, 0.4)0.68
    • Table 2. Training set and test set under standard operating condition

      View table

      Table 2. Training set and test set under standard operating condition

      ClassTraining set(17°)Test set(15°)
      Configuration(SN)Number of samplesConfiguration(SN)Number of samples
      BMP2956323395639566C21195196196
      BTR70233196
      T72132232132812S7196195191
      T62299273
      BRDM2298274
      BTR60256195
      ZSU23/4299274
      D7299274
      ZIL131299274
      2S1299274
    • Table 3. Average recognition rate under standard operating condition

      View table

      Table 3. Average recognition rate under standard operating condition

      MethodProposed methodMulti-resolution 1Multi-resolution 2MonogenicCNN
      Average recognition rate/%98.9298.1798.4397.8197.68
    • Table 4. Training set and test set under different pitch angles

      View table

      Table 4. Training set and test set under different pitch angles

      ClassTraining setTest set
      Pitch angleNumber of samplesPitch angleNumber of samples
      2S117°29930°45°288303
      BRDM217°29830°45°287303
      ZSU23/417°29930°45°288303
    • Table 5. Average recognition rate under different pitch angles

      View table

      Table 5. Average recognition rate under different pitch angles

      MethodAverage recognition rate/%
      30°45°
      Proposed method97.8674.17
      Multi-resolution 196.5271.26
      Multi-resolution 297.0771.82
      Monogenic96.4870.26
      CNN95.6865.56
    Tools

    Get Citation

    Copy Citation Text

    Jie He, Xin Chen. Multi-Resolution Selection of SAR Images and Target Recognition Based on Nonlinear Correlation Information Entropy[J]. Laser & Optoelectronics Progress, 2020, 57(22): 221020

    Download Citation

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

    Category: Image Processing

    Received: Mar. 26, 2020

    Accepted: Apr. 27, 2020

    Published Online: Nov. 10, 2020

    The Author Email: Jie He (sunflower_xpj@163.com)

    DOI:10.3788/LOP57.221020

    Topics