Journals Articles Conferences News About CLP
Submit a paper Email Alert
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
2D MaterialsTransformation opticsQuantum PhotonicsSmall lasersSemiconductor UV PhotonicsSupersymmetric microring laser arrays

Laser & Optoelectronics Progress, Volume. 57, Issue 16, 161025(2020)

Bronze Inscription Recognition Method Based on Automatic Pruning Strategy

Guoqiang Xia and Zhenhong Shang*
Author Affiliations
  • Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, Yunnan 650500, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (10)
    Equations (0)
    References (14)
    Tools
    Paper Information
    Topics
    Figures & Tables(10)
    Example of bronze inscription data set

    Fig. 1. Example of bronze inscription data set

    Download full size
    Training loss and accuracy curves on training and test data sets. (a) Accuracy curve; (b) loss curve

    Fig. 2. Training loss and accuracy curves on training and test data sets. (a) Accuracy curve; (b) loss curve

    Download full size
    L1 norm distributions of convolution kernes in C1,C2,C3 convolution layers . (a) C1 convolution layer; (b) C2 convolution layer; (c) C3 convolution layer

    Fig. 3. L1 norm distributions of convolution kernes in C1,C2,C3 convolution layers . (a) C1 convolution layer; (b) C2 convolution layer; (c) C3 convolution layer

    Download full size
    Final model retraining process and L1 norm distributions in convolution layers. (a) Retraining process; (b) L1 norm distributions

    Fig. 4. Final model retraining process and L1 norm distributions in convolution layers. (a) Retraining process; (b) L1 norm distributions

    Download full size
    L1 norm distributions of convolution kernels in VGG16 convolution layer. (a) Before pruning; (b) after downsizing; (c) after pruning

    Fig. 5. L1 norm distributions of convolution kernels in VGG16 convolution layer. (a) Before pruning; (b) after downsizing; (c) after pruning

    Download full size
    L1 norm distributions of convolution kernels in ResNet18 convolution layer. (a) Before pruning; (b) after pruning

    Fig. 6. L1 norm distributions of convolution kernels in ResNet18 convolution layer. (a) Before pruning; (b) after pruning

    Download full size

    • Table 1. Network structure of LeNet

      View table

      Table 1. Network structure of LeNet

      LayerInput sizeOutput sizeNumber of kernelsNumber of parametersFLOPS
      C132×32×130×30×666054000
      M130×30×615×15×65400
      C215×15×611×11×16162416292336
      M211×11×165×5×161936
      C35×5×161×1×32321283212832
      FC3210634983392
      Total18806369896

    • Table 2. Iterative process of pruning and retraining

      View table

      Table 2. Iterative process of pruning and retraining

      Number of iterationsC'1C'2C'3Number of parametersFLOPSAccuracy /%Accuracy after retraining /%
      06163218806369896100
      161315860929800382.36100
      261314817729757168.81100
      361313774529713962.93100
      461312731329670767.62100
      561311688129627565.7999.77
      661310644929584359.5298.71
      761210604827732251.8598.04
      861110564725880143.4697.62
      96119526525841939.989.17

    • Table 3. Convolution layer structures of VGG16 and ResNet18

      View table

      Table 3. Convolution layer structures of VGG16 and ResNet18

      NetworkNumber of kernelsFLOPS /106Number of parameters /103
      VGG1664(C1),64(C2),128(C3),128(C4),256(C5),256(C6),256(C7), 512(C8),512(C9),512(C10),512(C11),512(C12),512(C13)333.1934031
      ResNet1864(C1),64(C2),64(C3),64(C4),64(C5),128(C6),128(C7),128(C8),128(C9), 256(C10),256(C11),256(C12),256(C13),512(C14),512(C15),512(C16),512(C17)37.2211230

    • Table 4. Pruning results of VGG16 and ResNet18

      View table

      Table 4. Pruning results of VGG16 and ResNet18

      NetworkNumber ofiterationsNumber of kernels after pruningFLOPS /106Number ofparameters /103Accuracy /%
      VGG161839(C1),27(C2),86(C3),62(C4),93(C5),156(C6),0(C7), 176(C8),0(C9),0(C10),0(C11),0(C12),0(C13)62.291845077.96
      ResNet183964(C1),22(C2),64(C3),11(C4),64(C5),22(C6),128(C7),22(C8),128(C9),51(C10),256(C11),49(C12),256(C13),50(C14),512(C15),47(C16),512(C17)9.22154977.86
    Tools

    Get Citation

    Copy Citation Text

    Guoqiang Xia, Zhenhong Shang. Bronze Inscription Recognition Method Based on Automatic Pruning Strategy[J]. Laser & Optoelectronics Progress, 2020, 57(16): 161025

    Download Citation

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

    Category: Image Processing

    Received: Feb. 7, 2020

    Accepted: Mar. 19, 2020

    Published Online: Aug. 5, 2020

    The Author Email: Shang Zhenhong (shangzhenhong@126.com)

    DOI:10.3788/LOP57.161025

    Topics

    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology