• Laser & Optoelectronics Progress
  • Vol. 57, Issue 14, 141029 (2020)
Hong Zhang, Xinlan Zuo, and Yao Huang*
Author Affiliations
  • College of Computer and Information Technology, China Three Gorges University, Yichang, Hubei 443002, China
  • show less
    DOI: 10.3788/LOP57.141029 Cite this Article Set citation alerts
    Hong Zhang, Xinlan Zuo, Yao Huang. Feature Selection Based on the Correlation of Sparse Coefficient Vectors with Application to SAR Target Recognition[J]. Laser & Optoelectronics Progress, 2020, 57(14): 141029 Copy Citation Text show less
    Procedure of feature selection based on the correlation of sparse coefficient vectors and SAR target recognition
    Fig. 1. Procedure of feature selection based on the correlation of sparse coefficient vectors and SAR target recognition
    Optical images of the ten targets in the MSTAR dataset
    Fig. 2. Optical images of the ten targets in the MSTAR dataset
    Classification results under SOC
    Fig. 3. Classification results under SOC
    Average recognition rates under depression angle variance
    Fig. 4. Average recognition rates under depression angle variance
    ClassTraining setTest set
    BMP2233 (Sn_9563)195 (Sn_9563)
    BTR70233 (Sn_c71)196 (Sn_c71)
    T72232 (Sn_132)196 (Sn_132)
    T62299273
    BRDM2298274
    BTR60256195
    ZSU23/4299274
    D7299274
    ZIL131299274
    2S1299274
    Table 1. Training and test samples under SOC
    MethodAverage recognition rate /%
    Proposed method99.23
    Reference method 199.08
    Reference method 298.83
    Reference method 398.37
    Table 2. Average recognition rates under SOC
    SetBMP2BDRM2BTR70T72
    Trainingset233 (Sn_9563)298233232 (Sn_132)
    Test set428 (Sn_9566)429 (Sn_c21)00426 (Sn_812)573 (Sn_A04)573 (Sn_A05)573 (Sn_A07)567 (Sn_A10)
    Table 3. Training and test samples under configuration variance
    Test sampleClassification resultRecognition rate /%
    ClassConfigurationBMP2BRDM2BTR70T72
    BMP2Sn_956641383496.50
    Sn_c2141582496.74
    T72Sn_812115240895.77
    Sn_A0465855496.68
    Sn_A0527456097.73
    Sn_A0747455897.38
    Sn_A1071354996.83
    Table 4. Recognition results under configuration variance
    MethodAverage recognition rate /%
    Proposed method96.86
    Reference method 194.78
    Reference method 295.92
    Reference method 395.17
    Table 5. Average recognition rates under configuration variance
    SetDepression2S1BDRM2ZSU23/4
    Training set17°299298299
    Test set30°288287288
    45°303303303
    Table 6. Training and test samples under depression angle variance
    Hong Zhang, Xinlan Zuo, Yao Huang. Feature Selection Based on the Correlation of Sparse Coefficient Vectors with Application to SAR Target Recognition[J]. Laser & Optoelectronics Progress, 2020, 57(14): 141029
    Download Citation