• Infrared and Laser Engineering
  • Vol. 51, Issue 3, 20210421 (2022)
Jianhua Lu
Author Affiliations
  • School of Physics and Electronic Engineering, Yancheng Normal University, Yancheng 224007, China
  • show less
    DOI: 10.3788/IRLA20210421 Cite this Article
    Jianhua Lu. Decision fusion of CNN and SRC with application to SAR target recognition[J]. Infrared and Laser Engineering, 2022, 51(3): 20210421 Copy Citation Text show less
    Analysis of procedure of the recognition method
    Fig. 1. Analysis of procedure of the recognition method
    Confusion matrix for recognition of 10-class targets
    Fig. 2. Confusion matrix for recognition of 10-class targets
    Average recognition rates under experiment 4
    Fig. 3. Average recognition rates under experiment 4
    LayerConvolution/Pooling kernelSize of feature map
    Input88×88×1
    Convolution 15×5×2084×84×20
    Pooling 12×2×2042×42×20
    Convolution 25×5×4038×38×40
    Pooling 22×2×4019×19×40
    Convolution 34×4×8016×16×80
    Pooling 32×2×808×8×80
    Convolution 43×3×1606×6×160
    Pooling 42×2×1603×3×160
    Convolution 53×3×N1×1×N
    softmaxN
    Table 1. Descriptions of different layers in CNN
    Class TrainingTesting
    Elevation angle/(°)Sample amountElevation angle/(°)Sample amount
    BMP217 21415 174
    BTR70214175
    T72213175
    T62278256
    BRDM2277257
    BTR60234174
    ZSU23/4278249
    D7278249
    ZIL131278249
    2S1278249
    Table 2. Training and testing samples under experiment 1: Including 10-class targets
    Recognition methodOursSVMSRCCNN
    Average recognition rate99.36%98.64%98.23%99.08%
    Table 3. Average recognition rates under experiment 1
    Class TrainingTesting
    Elevation angle/(°)ConfigurationSample amountElevation angle/(°)ConfigurationSample amount
    BMP2 17 9 563 214 15 9 566175
    c21175
    BTR70c71214c71175
    T72 132 213 812174
    s7167
    Table 4. Training and testing samples under experiment 2: Including 3-class targets
    Recognition methodOursSVMSRCCNN
    Average recognition rate95.42%92.58%92.14%93.96%
    Table 5. Average recognition rates under experiment 2
    ClassTrainingTesting
    Elevation angle/(°)Sample amountElevation angle/(°)Sample amount
    2S11727730267
    45285
    BRDM227630266
    45285
    ZSU23/427730267
    45285
    Table 6. Training and testing samples under experiment 3: Including 3-class targets
    Recognition methodOursSVMSRCCNN
    Average recognition rate 30°97.56%94.52%95.87%97.04%
    45°71.64%66.64%65.42%67.56%
    Table 7. Average recognition rates under experiment 3
    Jianhua Lu. Decision fusion of CNN and SRC with application to SAR target recognition[J]. Infrared and Laser Engineering, 2022, 51(3): 20210421
    Download Citation