• Laser & Optoelectronics Progress
  • Vol. 58, Issue 4, 0410016 (2021)
Jiang Chang1, Shengqi Guan1、2、*, Hongyu Shi3, Luping Hu1, and Yiqi Ni1
Author Affiliations
  • 1School of Mechanical and Electronic Engineering, Xi'an Polytechnic University, Xi'an, Shaanxi 710048, China
  • 2Shaoxing Keqiao West-Tex Textile Industry Innovative Institute, Shaoxing, Zhejiang 312030, China
  • 3School of Computer Science, Xi'an Polytechnic University, Xi'an, Shaanxi 710048, China
  • show less
    DOI: 10.3788/LOP202158.0410016 Cite this Article Set citation alerts
    Jiang Chang, Shengqi Guan, Hongyu Shi, Luping Hu, Yiqi Ni. Strip Defect Classification Based on Improved Generative Adversarial Networks and MobileNetV3[J]. Laser & Optoelectronics Progress, 2021, 58(4): 0410016 Copy Citation Text show less
    Structure of the generator network
    Fig. 1. Structure of the generator network
    Structure of the discriminator network
    Fig. 2. Structure of the discriminator network
    Part of the original strip steel defect images
    Fig. 3. Part of the original strip steel defect images
    Images obtained by different network training. (a) 100 epochs; (b) 300 epochs; (c) 500 epochs
    Fig. 4. Images obtained by different network training. (a) 100 epochs; (b) 300 epochs; (c) 500 epochs
    OperatorResolutionChannelLayer
    Conv 3×364×64161
    Bottleneck 3×332×32242
    Bottleneck 5×516×16402
    Bottleneck 5×58×8802
    Conv 1×1 & Pooling4×41201
    Conv 1×11×12401
    Conv 1×11×161
    Table 1. Structure of the improved MobileNetV3
    NetworkParameter amount /millionPredictedtime /msAccuracy /%Precision /%Recall /%F1
    Original MobileNetV31.2427292.0592.5491.470.92
    Improved MobileNetV30.0923194.6795.2194.790.95
    VGG1617.3532093.8594.3393.670.94
    ResNet3421.2831891.2391.4190.590.91
    DenseNet1216.9531594.3694.5094.100.94
    Table 2. Image classification performance of different networks
    Jiang Chang, Shengqi Guan, Hongyu Shi, Luping Hu, Yiqi Ni. Strip Defect Classification Based on Improved Generative Adversarial Networks and MobileNetV3[J]. Laser & Optoelectronics Progress, 2021, 58(4): 0410016
    Download Citation