Journal of Applied Optics, Volume. 43, Issue 1, 87(2022)

Defects detection method of photovoltaic cells based on lightweightconvolutional neural network

Huaiguang LIU1...2, Wancheng DING1,* and Qianwen HUANG1 |Show fewer author(s)
Author Affiliations
  • 1Key Laboratory of Metallurgical Equipment and Control Technology (Ministry of Education), Wuhan University of Science and Technology, Wuhan 430081, China
  • 2Institute of Robotics and Intelligent Systems, Wuhan University of Science and Technology, Wuhan 430081, China
  • show less
    Figures & Tables(15)
    Imaging principle of PL technology
    Cell and PL image
    Types of cell defects to be detected
    Network structure of YOLOv3-Tiny
    Improved cell defects detection network structure
    Fusion of shallow feature enhancement layers
    Schematic diagram of bounding box prediction
    Partial data set after data enhancement
    Curves of loss function
    Comparison of PR curves
    Comparison of detection results before and after model improvement
    Online detection site
    • Table 1. Division of data set

      View table
      View in Article

      Table 1. Division of data set

      数据集第1种隐裂/张第2种隐裂/张划痕/张黑斑/张总数量/张
      训练集2 8003 8002 4004 00012 000
      验证集7508006308203 000
      测试集6808206608403 000
    • Table 2. Comparison of AP values of two network models

      View table
      View in Article

      Table 2. Comparison of AP values of two network models

      网络模型第1种隐裂第2种隐裂划痕黑斑
      YOLOv3-Tiny的AP值/%79.1990.0064.1089.80
      本文网络的AP值/%83.0492.6880.8393.66
    • Table 3. Performance comparison of different network models

      View table
      View in Article

      Table 3. Performance comparison of different network models

      网络模型TmAP/% FPS
      SSD97.5212
      YOLOv394.3125
      YOLOv3-Tiny80.7747
      本文网络87.5540
    Tools

    Get Citation

    Copy Citation Text

    Huaiguang LIU, Wancheng DING, Qianwen HUANG. Defects detection method of photovoltaic cells based on lightweightconvolutional neural network[J]. Journal of Applied Optics, 2022, 43(1): 87

    Download Citation

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

    Category: OPTICAL METROLOGY AND MEASUREMENT

    Received: Aug. 18, 2021

    Accepted: --

    Published Online: Mar. 7, 2022

    The Author Email: DING Wancheng (dingwancheng_wust@163.com)

    DOI:10.5768/JAO202243.0103003

    Topics