Chinese Optics, Volume. 17, Issue 4, 823(2024)

Detection of surface and internal defects in cold rolled steel

Ming-yu CHEN1,2, Yue-chen XIE1,2, Xiong-tao LV1,2, Jian-rong GUO3, Guo-jun JIA4, Zhi-peng XU2, Shi-ling WANG5, Zhen XIANG1, and Dong LIU1,2、*
Author Affiliations
  • 1State Key Laboratory of Extreme Photonics and Instrumentation, College of Optical Science and Engineering, Zhejiang University, Hangzhou 310027, China
  • 2ZJU-Hangzhou Global Scientific and Technological Innovation Center, Hangzhou 311200, China
  • 3Zhejiang Southeast New Material Technology Company Limited, Hangzhou 311222, China
  • 4Technology Center, Ningbo Steel Company Limited, Ningbo 315807, China
  • 5College of Metrology and Measurement Engineering, China Jiliang University, Hangzhou 310018, China
  • show less
    Figures & Tables(17)
    Cold rolled steel defect detection scheme
    Model of cold rolled steel surface detection
    Detecting scratches captured at different angles by camera不同相机拍摄角下的检测划痕
    Lighting direction of three kinds of line light sources
    Simulation results of scratches illuminated by different light sources
    The cold rolled steel surface defect detection system
    Sample with blind hole
    Defect identification results
    Precision-Recall curve for different defects under white line light source and white bilateral line light source illumination. (a) Inclusion; (b) tree grain; (c) scratch
    Variation of mAP:0.5 with the epochs under white line light source and white bilateral line light source illumination
    Detection results of internal defects
    Geometric magnification principle for X-ray detection
    Blind hole sample edge detection using Sobel algorithm
    • Table 1. Defect size in cold rolled steel samples

      View table
      View in Article

      Table 1. Defect size in cold rolled steel samples

      缺陷类型长度均值/cm不确定度宽度均值/cm不确定度
      孔洞0.8960.0350.5120.020
      破损4.9600.0251.2040.030
      夹杂15.520.740.1620.091
      树纹8.2420.0300.2000.025
      划痕20.490.740.02220.0084
      色差19.700.745.9940.030
    • Table 2. Imaging comparison when using white lateral and white bilateral line light source illuminations

      View table
      View in Article

      Table 2. Imaging comparison when using white lateral and white bilateral line light source illuminations

      孔洞破损夹杂树纹纵向划痕横向划痕色差
      白光单侧线光源
      白光双侧线光源
    • Table 3. Results of YOLOv5 target detection algorithm using white line light source and white bilateral line light source illuminations

      View table
      View in Article

      Table 3. Results of YOLOv5 target detection algorithm using white line light source and white bilateral line light source illuminations

      白光线光源白光双侧线光源
      准确率80.80%91.50%↑
      召回率96.00%97.67%↑
      mAP:0.574.70%90.16%↑
      损失值1.38%1.37%
      孔洞99.50%99.50%
      破损99.60%99.60%
      夹杂74.80%89.80%↑
      树纹74.60%96.27%↑
      划痕65.57%83.90%↑
      色差\73.00%
    • Table 4. Results of edge detection on blind hole sample by applying the Sobel algorithm

      View table
      View in Article

      Table 4. Results of edge detection on blind hole sample by applying the Sobel algorithm

      工业CT超声检测红外热波成像
      盲孔检出数202015
      盲孔边缘连通数201913
      边缘提取准确数201910
      最大盲孔边缘灰度差值
      φ2.50 mm190193104
      φ1.75 mm18718080
      φ1.00 mm18417230
      φ0.25 mm14589/
    Tools

    Get Citation

    Copy Citation Text

    Ming-yu CHEN, Yue-chen XIE, Xiong-tao LV, Jian-rong GUO, Guo-jun JIA, Zhi-peng XU, Shi-ling WANG, Zhen XIANG, Dong LIU. Detection of surface and internal defects in cold rolled steel[J]. Chinese Optics, 2024, 17(4): 823

    Download Citation

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

    Received: Oct. 31, 2023

    Accepted: --

    Published Online: Aug. 9, 2024

    The Author Email:

    DOI:10.37188/CO.2023-0189

    Topics