留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

冷轧钢表面与内部缺陷检测研究

陈名渝 谢玥辰 吕雄涛 郭建荣 贾国军 许志鹏 王狮凌 项震 刘东

陈名渝, 谢玥辰, 吕雄涛, 郭建荣, 贾国军, 许志鹏, 王狮凌, 项震, 刘东. 冷轧钢表面与内部缺陷检测研究[J]. 中国光学(中英文). doi: 10.37188/CO.2023-0189
引用本文: 陈名渝, 谢玥辰, 吕雄涛, 郭建荣, 贾国军, 许志鹏, 王狮凌, 项震, 刘东. 冷轧钢表面与内部缺陷检测研究[J]. 中国光学(中英文). doi: 10.37188/CO.2023-0189
CHEN Ming-yu, XIE Yue-chen, LV Xiong-tao, GUO Jian-rong, JIA Guo-jun, XU Zhi-peng, WANG Shi-ling, XIANG Zhen, LIU Dong. Research on the detection of surface and internal defects in cold rolled steel[J]. Chinese Optics. doi: 10.37188/CO.2023-0189
Citation: CHEN Ming-yu, XIE Yue-chen, LV Xiong-tao, GUO Jian-rong, JIA Guo-jun, XU Zhi-peng, WANG Shi-ling, XIANG Zhen, LIU Dong. Research on the detection of surface and internal defects in cold rolled steel[J]. Chinese Optics. doi: 10.37188/CO.2023-0189

冷轧钢表面与内部缺陷检测研究

doi: 10.37188/CO.2023-0189
基金项目: 国家重点研发计划(No. 2022YFB3403404);极端光学技术与仪器全国重点实验室创新项目(No. EPI2023ZD01)
详细信息
    作者简介:

    陈名渝(1999—),女,中国台湾人,硕士研究生在读,2021年于中山大学获得学士学位,主要研究方向为光学检测。E-mail:22130122@zju.edu.cn

    刘 东(1982—),男,教授,博士,博士生导师,分别于2005年和2010年在浙江大学获得学士和博士学位,曾在美国宇航局(NASA)从事博士后研究工作。主要研究方向为光学检测、激光雷达、机器视觉、深度学习。E-mail: liudongopt@zju.edu.cn

  • 中图分类号: O439

Research on the detection of surface and internal defects in cold rolled steel

Funds: Supported by National key research and development program (No. 2022YFB3403404); State Key Laboratory of Extreme Photonics and Instrumentation Innovation Program (No. EPI2023ZD01)
More Information
  • 摘要:

    为实现冷轧钢缺陷全面检测,本文针对其表面和内部缺陷检测进行研究。表面缺陷检测中,提出双侧线光源照明方案,并与常规线光源作对比。内部缺陷检测中,从检测分辨率和缺陷边缘特征两方面,分析X射线、超声以及红外热波成像等金属内部检测技术的适用性。经实验验证,双侧线光源照明不仅使YOLOv5目标检测算法模型总体平均精度mAP:0.5达到90.16%,相比线光源提升15.46%,还可优化模型分类和提高训练效率。X射线和超声波检测法可检测直径0.25 mm盲孔,而红外热波成像技术则可有效识别出直径1 mm盲孔。在缺陷边缘特征评估中,X射线检测法的最小盲孔边缘灰度差值为145,超声波为89,红外热波成像为30。本研究为冷轧钢表面缺陷检测改进需求,并为内部缺陷检测研究提供了思路。

     

  • 图 1  冷轧钢缺陷检测方案

    Figure 1.  Cold rolled steel defect detection scheme

    图 2  冷轧钢表面检测模型

    Figure 2.  Model of cold rolled steel surface detection

    图 3  不同相机拍摄角${\theta _r}$检测划痕

    Figure 3.  Detecting scratches at different camera angles ${\theta _r}$

    图 4  线光源照明方向

    Figure 4.  Lighting direction of line light sources

    图 5  不同光源照明各方向划痕仿真

    Figure 5.  Simulation of scratches with different light sources

    图 6  冷轧钢表面缺陷检测模块

    Figure 6.  The cold rolled steel surface defect detection system

    图 7  盲孔加工样品

    Figure 7.  Sample with blind hole

    图 8  样品缺陷识别结果

    Figure 8.  Defect identification result

    图 9  白光线光源与白光双侧线光源照明下Precision-Recall曲线

    Figure 9.  Precision-Recall curve with white line light source and white bilateral line light source illumination

    图 10  白光线光源与白光双侧线光源照明下平均精度mAP:0.5随训练轮数变化关系

    Figure 10.  Variation of mAP:0.5 with the epochs between white line light source and white bilateral line light source illumination

    图 11  内部缺陷检测结果

    Figure 11.  Detection results of internal defects

    图 12  X射线检测几何放大原理

    Figure 12.  Geometric magnification principle for X-ray detection

    图 13  Sobel算法盲孔样品边缘特征提取

    Figure 13.  Sobel algorithm blind hole sample edge detection

    表  1  冷轧钢样品缺陷尺寸

    Table  1.   Size of defects 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 0.896 0.035 0.512 0.020
    破损 4.960 0.025 1.204 0.030
    夹杂 15.52 0.74 0.162 0.091
    树纹 8.242 0.030 0.200 0.025
    划痕 20.49 0.74 0.0222 0.0084
    色差 19.70 0.74 5.994 0.030
    下载: 导出CSV

    表  2  白光线光源与白光双侧线光源照明成像对比

    Table  2.   Imaging comparison of white and white bilateral line light source illumination

    孔洞破损夹杂树纹纵向划痕横向划痕色差
    白光
    线光源
    白光
    双侧线光源
    下载: 导出CSV

    表  3  白光线光源与白光双侧线光源照明下YOLOv5目标检测算法结果

    Table  3.   Results of YOLOv5 Target Detection Algorithm with white line light source and white bilateral line light source illumination

    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:0.5 74.70% 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%
    下载: 导出CSV

    表  4  Sobel算法盲孔样品边缘特征提取结果

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

    工业CT 超声检测 红外热波成像
    盲孔检出数 20 20 15
    盲孔边缘连通数 20 19 13
    边缘提取准确数 20 19 10
    最大盲孔边缘灰度差值
    φ2.50 mm 190 193 104
    φ1.75 mm 187 180 80
    φ1.00 mm 184 172 30
    φ0.25 mm 145 89 /
    下载: 导出CSV
  • [1] LI ZH, LI ZH, HUANG H, et al. Green lithium: photoelectrochemical extraction[J]. PhotoniX, 2023, 4(1): 23.
    [2] ZHU Y N, LUO H, YANG CH Y, et al. Color-preserving passive radiative cooling for an actively temperature-regulated enclosure[J]. Light: Science & Applications, 2022, 11(1): 122.
    [3] 李玮. 用于锂离子电池钢壳的镀镍钴钢带制备工艺及耐蚀性研究[D]. 湘潭: 湘潭大学, 2011.

    LI W. The study of preparation and corrosion resistance of cobalt-nickel-plated steel strip for lithium-ion battery shell[D]. Xiangtan: Xiangtan University, 2011. (in Chinese).
    [4] 赵小龙, 王雄, 罗晓阳, 等. 退火温度对CSP冷轧低碳高强钢带组织和性能的影响[J]. 金属热处理,2021,46(1):43-49.

    ZHAO X L, WANG X, LUO X Y, et al. Effect of annealing temperature on microstructure and properties of cold rolled low carbon high strength sheet by CSP[J]. Heat Treatment of Metals, 2021, 46(1): 43-49. (in Chinese).
    [5] MANCINI E, CAMPANA F, SASSO M, et al. Effects of cold rolling process variables on final surface quality of stainless steel thin strip[J]. The International Journal of Advanced Manufacturing Technology, 2012, 61(1-4): 63-72.
    [6] 董双鹏, 宋进英, 陈业熊, 等. 冷轧钢板表面缺陷的种类及形成原因[J]. 机械工程材料,2012,36(7):100-104.

    DONG SH P, SONG J Y, CHEN Y X, et al. Types and forming reasons of cold rolled steel plate surface defects[J]. Materials for Mechanical Engineering, 2012, 36(7): 100-104. (in Chinese).
    [7] 胡华东, 王乐, 汤化胜, 等. 冷轧宽厚规格深冲钢梗印缺陷成因分析及其控制措施[J]. 轧钢,2022,39(2):124-128.

    HU H D, WANG L, TANG H SH, et al. Cause analysis of stem seal defects of cold rolled wide and thick gauge deep drawing steel strip and its control measures[J]. Steel Rolling, 2022, 39(2): 124-128. (in Chinese).
    [8] 王畅, 于洋, 王林, 等. 冷轧双相钢镀锌条带状色差缺陷产生原因及机理研究[J]. 轧钢,2023,40(1):52-58.

    WANG CH, YU Y, WANG L, et al. Study on the cause and mechanism of the strip color difference defects of cold rolled dual- phase steel galvanized strip[J]. Steel Rolling, 2023, 40(1): 52-58. (in Chinese).
    [9] SURESH B R, FUNDAKOWSKI R A, LEVITT T S, et al. A real-time automated visual inspection system for hot steel slabs[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1983, PAMI-5(6): 563-572.
    [10] NATH V, CHATTOPADHYAY C, DESAI K A. On enhancing prediction abilities of vision-based metallic surface defect classification through adversarial training[J]. Engineering Applications of Artificial Intelligence, 2023, 117: 105553.
    [11] 杨水山, 何永辉, 王康健. 高速冷连轧带钢表面质量在线检测系统开发与应用[J]. 世界钢铁,2013,13(3):62-67. doi: 10.3969/j.issn.1672-9587.2013.03.011

    YANG SH SH, HE Y H, WANG K J. Development on on-line inspection system for high speed strip surface quality at cold continuous rolling mill[J]. World Iron & Steel, 2013, 13(3): 62-67. (in Chinese). doi: 10.3969/j.issn.1672-9587.2013.03.011
    [12] 沈春光, 李虎威, 荆涛, 等. 基于深度学习的带钢表面缺陷检测在小样本数据集的应用[J]. 轧钢,2022,39(2):82-86.

    SHEN CH G, LI H W, JING T, et al. Application of deep learning-based surface defect recognition of plate in small sample dataset[J]. Steel Rolling, 2022, 39(2): 82-86. (in Chinese).
    [13] JUNet A, MESSAGER A, WECK A, et al. Internal fatigue crack propagation in a Ti-6Al-4V alloy: an in situ study[J]. International Journal of Fatigue, 2023, 168: 107450.
    [14] YANG Q ZH, ZHAO Y, HU X A, et al. Small crack propagation and closure behaviours and mechanisms in a powder metallurgy superalloy at high temperature in air[J]. International Journal of Fatigue, 2023, 175: 107797.
    [15] 白贇沨, 刘祎, 张小琳, 等. 基于GF-WLS和VSM的多能量X射线图像融合方法[J]. 无损检测,2022,44(9):34-41,74. doi: 10.11973/wsjc202209007

    BAI Y F, LIU Y, ZHANG X L, et al. Multi energy X-ray image fusion method based on GF-WLS and VSM[J]. Nondestructive Testing, 2022, 44(9): 34-41,74. (in Chinese). doi: 10.11973/wsjc202209007
    [16] TAKADA H, TOMURA Y, ARATANI M, et al. On-line detection system for internal flaws in as-hot-rolled steel strip using ultrasonic probe array[J]. Materials Transactions, 2011, 52(3): 531-538.
    [17] 张永生, 黄松岭, 赵伟, 等. 基于电磁超声的钢板裂纹检测系统[J]. 无损检测,2009,31(4):307-310.

    ZHANG Y SH, HUANG S L, ZHAO W, et al. Detection system about steel cracks based on electronic ultrasonic[J]. Nondestructive Testing, 2009, 31(4): 307-310. (in Chinese).
    [18] LIU SH Y, JIA K N, WAN H P, et al. Inspection of the internal defects with different size in Ni and Ti additive manufactured components using laser ultrasonic technology[J]. Optics & Laser Technology, 2022, 146: 107543.
    [19] ALMOND D P, PICKERING S G. An analytical study of the pulsed thermography defect detection limit[J]. Journal of Applied Physics, 2012, 111(9): 093510.
    [20] ZHUO L J, YANG X, ZHU J G, et al. Size determination of interior defects by reconstruction of subsurface virtual heat flux for step heating thermography[J]. NDT & E International, 2023, 133: 102734.
    [21] WANG F, SHENG J, SFARRA S, et al. Multimode infrared thermal-wave imaging in non-destructive testing and evaluation (NDT&E): physical principles, modulated waveform, and excitation heat source[J]. Infrared Physics & Technology, 2023, 135: 104993.
    [22] NAYAR S K, IKEUCHI K, KANADE T. Surface reflection: physical and geometrical perspectives[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1991, 13(7): 611-634.
    [23] 张青树, 齐韦. 冷轧带钢表面形貌测试及其粗糙度均匀性控制研究[J]. 轧钢,2022,39(5):103-107.

    ZHANG Q SH, QI W. Study on surface profile test and roughness uniformity control of cold rolled strip[J]. Steel Rolling, 2022, 39(5): 103-107. (in Chinese).
    [24] 杨甬英, 陆春华, 梁蛟, 等. 光学元件表面缺陷的显微散射暗场成像及数字化评价系统[J]. 光学学报,2007,27(6):1031-1038. doi: 10.3321/j.issn:0253-2239.2007.06.015

    YANG Y Y, LU CH H, LIANG J, et al. Microscopic dark-field scattering imaging and digitalization evaluation system of defects on optical devices precision surface[J]. Acta Optica Sinica, 2007, 27(6): 1031-1038. (in Chinese). doi: 10.3321/j.issn:0253-2239.2007.06.015
    [25] LI X C, WU Q Q, XIAO B P, et al. High-speed and robust infrared-guiding multiuser eye localization system for autostereoscopic display[J]. Applied Optics, 2020, 59(14): 4199-4208.
    [26] 吴海滨, 魏喜盈, 刘美红, 等. 结合空洞卷积和迁移学习改进YOLOv4的X光安检危险品检测[J]. 中国光学,2021,14(6):1417-1425.

    WU H B, WEI X Y, LIU M H, et al. Improved YOLOv4 for dangerous goods detection in X-ray inspection combined with atrous convolution and transfer learning[J]. Chinese Optics, 2021, 14(6): 1417-1425. (in Chinese).
    [27] 张寅, 朱桂熠, 施天俊, 等. 基于特征融合与注意力的遥感图像小目标检测[J]. 光学学报,2022,42(24):2415001.

    ZHANG Y, ZHU G Y, SHI T J, et al. Small object detection in remote sensing images based on feature fusion and attention[J]. Acta Optica Sinica, 2022, 42(24): 2415001. (in Chinese).
    [28] 张印辉, 庄宏, 何自芬, 等. 氨气泄漏混洗自注意力轻量化红外检测[J]. 中国光学(中英文),2023,16(3):607-619.

    ZHENG Y H, ZHUANG H, HE Z F, et al. Lightweight infrared detection of ammonia leakage using shuffle and self-attention[J]. Chinese Optics, 2023, 16(3): 607-619. (in Chinese).
    [29] 刘彦磊, 李孟喆, 王宣宣. 轻量型YOLOv5s车载红外图像目标检测[J]. 中国光学(中英文),2023,16(5):1045-1055.

    LIU Y L, LI M ZH, WANG X X. Lightweight YOLOv5s vehicle infrared image target detection[J]. Chinese Optics, 2023, 16(5): 1045-1055. (in Chinese).
    [30] VILLARRAGA-GÓMEZ H, SMITH S T. Effect of geometric magnification on dimensional measurements with a metrology-grade X-ray computed tomography system[J]. Precision Engineering, 2022, 73: 488-503.
  • 加载中
图(13) / 表(4)
计量
  • 文章访问数:  52
  • HTML全文浏览量:  32
  • PDF下载量:  13
  • 被引次数: 0
出版历程
  • 网络出版日期:  2024-05-09

目录

    /

    返回文章
    返回

    重要通知

    2024年2月16日科睿唯安通过Blog宣布,2024年将要发布的JCR2023中,229个自然科学和社会科学学科将SCI/SSCI和ESCI期刊一起进行排名!《中国光学(中英文)》作为ESCI期刊将与全球SCI期刊共同排名!