• Laser & Optoelectronics Progress
  • Vol. 58, Issue 24, 2428004 (2021)
Kangjie Zheng1, Shan Jin2, and ChengWei Zhang2、*
Author Affiliations
  • 1Navigation College, Dalian Maritime University, Dalian, Liaoning 116026, China
  • 2School of Information Science and Technology, Dalian Maritime University, Dalian, Liaoning 116026, China
  • show less
    DOI: 10.3788/LOP202158.2428004 Cite this Article Set citation alerts
    Kangjie Zheng, Shan Jin, ChengWei Zhang. Research on Adversarial Examples in Human Physical Rehabilitation Exercises Based on GPREGAN Framework[J]. Laser & Optoelectronics Progress, 2021, 58(24): 2428004 Copy Citation Text show less
    GPREGAN framework
    Fig. 1. GPREGAN framework
    Network structure of generator
    Fig. 2. Network structure of generator
    Recognition rate of depth model. (a) CNN recognition rate; (b) LSTM recognition rate
    Fig. 3. Recognition rate of depth model. (a) CNN recognition rate; (b) LSTM recognition rate
    Recognition rate of depth model for adversarial examples. (a) CNN recognition rate; (b) LSTM recognition rate
    Fig. 4. Recognition rate of depth model for adversarial examples. (a) CNN recognition rate; (b) LSTM recognition rate
    Distance between adversarial example and original sample. (a) CNN; (b) LSTM
    Fig. 5. Distance between adversarial example and original sample. (a) CNN; (b) LSTM
    Recognition rate for adversarial examples. (a) CNN; (b) LSTM
    Fig. 6. Recognition rate for adversarial examples. (a) CNN; (b) LSTM
    SNR-8 dB-6 dB-4 dB-2 dB02 dB4 dB6 dB8 dB
    CNN0.76250.81750.91000.93000.96000.97250.97750.98500.9875
    LSTM0.26000.36250.44750.53750.72750.75250.81000.87000.8925
    Table 1. Influence of SNR on depth model
    Group No.12345678
    CNN0.01630.01540.01470.01390.01320.01270.01220.0119
    LSTM0.00080.01680.01080.01270.00490.01340.00470.0175
    Table 2. Mean square error between original sample and adversarial example
    Kangjie Zheng, Shan Jin, ChengWei Zhang. Research on Adversarial Examples in Human Physical Rehabilitation Exercises Based on GPREGAN Framework[J]. Laser & Optoelectronics Progress, 2021, 58(24): 2428004
    Download Citation