基于纹理特征与集成学习的热轧工作辊表面磨损检测方法研究

doi:10.3901/JME.260178

机械工程学报 ›› 2025, Vol. 62 ›› Issue (6): 111-120.doi: 10.3901/JME.260178

• 特邀专栏：轧制技术与智能化 • 上一篇

扫码分享

基于纹理特征与集成学习的热轧工作辊表面磨损检测方法研究

胡淇伟^1,2, 孙永吉^1,2, 张建超^1,2, 任新意³, 高慧敏³, 季策^1,2, 黄华贵^1,2

1. 燕山大学机械工程学院秦皇岛 066004;
2. 燕山大学国家冷轧板带装备及工艺工程技术研究中心秦皇岛 066004;
3. 首钢京唐钢铁联合有限公司技术中心唐山 063009

收稿日期:2025-08-10 修回日期:2025-12-10 发布日期:2026-05-12
作者简介:胡淇伟,男,1993年出生,博士研究生。主要研究方向为机器视觉检测与智能运维算法开发及应用。E-mail:huqiwei@stumail.ysu.edu.cn
黄华贵(通信作者),男,1978年出生,博士,教授,博士研究生导师。主要研究方向为双辊薄带铸轧、金属层状复合材料、机器视觉检测、冶金装备智能运维。E-mail:hhg@ysu.edu.cn
基金资助:
河北省高等学校科学技术研究(CXY2024009)、中央引导地方科技发展资金(216Z1602G)、河北省自然科学基金燕赵青年科学家(E2023203260)、河北省科技重大专项(23281901Z)、邢台市重大科技专项(2023ZZ017)和河北省创新能力提升计划(24431801D)资助项目。

Research on Surface Wear Detection Method for Hot Rolling Work Roll Based on Texture Features and Ensemble Learning

HU Qiwei^1,2, SUN Yongji^1,2, ZHANG Jianchao^1,2, REN Xinyi³, GAO Huimin³, JI Ce^1,2, HUANG Huagui^1,2

1. College of Mechanical Engineering, Yanshan University, Qinhuangdao 066004;
2. National Engineering Research Center for Equipment and Technology of Cold Strip Rolling, Yanshan University, Qinhuangdao 066004;
3. Shougang Jingtang Iron and Steel United Co., Ltd., Tangshan 063009

Received:2025-08-10 Revised:2025-12-10 Published:2026-05-12

摘要/Abstract

摘要： 热轧工作辊表层磨损状况的精准判定，对优化轧机效能具有显著工程价值。针对大尺寸辊面的成像难题，设计了融合双线阵相机与定制线光源的视觉采集系统，在换辊过程中稳定获取辊面图像。提出了基于约束动态时间规整的模板匹配算法，实现图像自动化采集。通过粗糙度、对比度、线性度与规则度构成的四维Tamura纹理特征向量有效表征了辊面磨损状态，结合极端随机树集成学习模型，对典型磨损形貌进行分类，准确率达95%。集成上述方法开发了辊面磨损智能检测平台，为热轧工作辊智能运维及工艺优化提供了新途径。

关键词: 热轧工作辊, 表面磨损评估, 机器视觉, 纹理特征提取, 极端随机树

Abstract: Precise determination of hot rolling work roll surface wear conditions is critical for optimizing mill performance. To address the imaging challenges posed by large-sized roll surfaces, a vision acquisition system integrating dual line-scan cameras with customized linear light sources was designed, allowing roll surface images to be reliably captured during the roll changing process. A template matching algorithm based on Constrained Dynamic Time Warping is proposed to achieve automated image acquisition. Wear conditions are effectively characterized by a four-dimensional Tamura texture feature vector consisting of coarseness, contrast, linearity, and regularity. Furthermore, the classification of typical wear morphologies is achieved with 95% accuracy by employing an extremely randomized trees ensemble learning model. Finally, an intelligent roll surface wear detection platform was developed, providing a novel approach for the intelligent maintenance and process optimization of hot rolling work rolls.

Key words: hot rolling work rolls, surface wear assessment, machine vision, texture feature extraction, extreme random trees

中图分类号:

TG335

胡淇伟, 孙永吉, 张建超, 任新意, 高慧敏, 季策, 黄华贵. 基于纹理特征与集成学习的热轧工作辊表面磨损检测方法研究[J]. 机械工程学报, 2025, 62(6): 111-120.

HU Qiwei, SUN Yongji, ZHANG Jianchao, REN Xinyi, GAO Huimin, JI Ce, HUANG Huagui. Research on Surface Wear Detection Method for Hot Rolling Work Roll Based on Texture Features and Ensemble Learning[J]. Journal of Mechanical Engineering, 2025, 62(6): 111-120.

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: http://www.cjmenet.com.cn/CN/10.3901/JME.260178

http://www.cjmenet.com.cn/CN/Y2025/V62/I6/111

参考文献

[1] 彭艳，石宝东，刘才溢，等. 板带轧制装备-工艺-产品质量综合控制融合发展综述[J]. 机械工程学报，2023，59(20)：96-118.

PENG Yan，SHI Baodong，LIU Caiyi，et al. Review of the integrated development of strip rolling equipment-process-product quality control[J]. Journal of Mechanical Engineering，2023，59(20)：96-118.

[2] BOMBAČ D，KUGLER G，BURJA J，et al. Early spalling analysis of large particles in high-Cr steel during thermal fatigue：Relevant mechanisms[J]. Materials，2022，15(19)：6705.

[3] JIANG Y，LI C，ZHANG X，et al. Surface defect detection of high precision cylindrical metal parts based on machine vision[M]. Cham：Springer International Publishing，2021.

[4] YE M，ZHANG W，ZHANG W，et al. Surface defects inspection of cylindrical metal workpieces based on weakly supervised learning[J]. International Journal of Advanced Manufacturing Technology，2022，119(3)：1933-1949.

[5] LUO Q，FANG X，LIU L，et al. Automated visual defect detection for flat steel surface：A survey[J]. IEEE Transactions on Instrumentation and Measurement，2020，69(3)：626-644.

[6] IBRAHIM A A M S，TAPAMO J R. A survey of vision - based methods for surface defects’ detection and classification in steel products[J]. Informatics，2024，11(2)：25.

[7] 伍麟，郝鸿宇，宋友. 基于计算机视觉的工业金属表面缺陷检测综述[J]. 自动化学报，2024，50(7)：1261-1283.

WU Lin，HAO Hongyu，SONG You. A review of metal surface defect detection based on computer vision[J]. Acta Automatica Sinica，2024，50(7)：1261-1283.

[8] 傅子秋，张晓龙，余成，等. 多场景下基于快速相机标定的柱面图像拼接方法[J]. 光电工程，2020，47(4)：74-86.

FU Ziqiu，ZHANG Xiaolong，YU Chen，et al. Cylindrical image mosaic method based on fast camera calibration in multi-scene [J]. Opto-Electronic Engineering，2020，47(4)：74-86.

[9] 赵春华，贾玥泽，谭金铃，等. 基于机器视觉的轴承外圈表面图像拼接方法[J]. 科学技术与工程，2022，22(14)：5729-5736.

ZHAO Chunhua，JIA Yueze，TAN Jinling，et al. Image mosaic method of bearing outer ring surface based on machine vision[J]. Science Technology and Engineering，2022，22(14)：5729-5736.

[10] 徐科，杨朝霖，周鹏. 热轧带钢表面缺陷在线检测的方法与工业应用[J]. 机械工程学报，2009，45(4)：111-114，124.

XU Ke，YANG Chaolin，ZHOU Peng. Technology of on-line surface inspection for hot-rolled steel strips and its industrial application[J]. Journal of Mechanical Engineering，2009，45(4)：111-114，124.

[11] SHAHABI H H，RATNAM M M. Noncontact roughness measurement of turned parts using machine vision[J]. International Journal of Advanced Manufacturing Technology，2010，46(1)：275-284.

[12] LIU B，YANG Y，WANG S，et al. An automatic system for bearing surface tiny defect detection based on multi-angle illuminations[J]. Optik，2020，208：164517.

[13] PHOA V，RAKHMAD H，PURWADI A. Binary adjacent set occurrence for a simple texture analysis in binary image[C]//International Conference on Computer Science，Information Technology，and Electrical Engineering. Jember，Indonesia：IEEE，2019：107-113.

[14] KALIST V，GANESAN P，LEO JOSEPH L M I，et al. Image quality analysis based on texture feature extraction using second-order statistical approach[C]//Innovations in Electronics and Communication Engineering. Singapore：Springer Singapore，2020：495-504.

[15] KLORIN G，HAYAT N，LINDER R，et al. Fourier transformation based texture analysis for differentiating between hyperplastic polyps and sessile serrated adenomas[J]. Microscopy Research and Technique，2023，86(4)：473-480.

[16] CAO M，MING D，XU L，et al. Frequency spectrum-based optimal texture window size selection for high spatial resolution remote sensing image analysis[J]. Journal of Spectroscopy，2019，2019：1-15.

[17] GARG M，JOYCE JACOB J，TUSHA，et al. Texture based image and video analysis[C]//International Conference on Computing Communication and Networking Technologies. Kamand，India：IEEE，2024：1-6.

[18] SONG G，SONG K，YAN Y. Saliency detection for strip steel surface defects using multiple constraints and improved texture features[J]. Optics and Lasers in Engineering，2020，128：106000.

[19] BOUDANI F Z，NACEREDDINE N，LAICHE N. Texture features extraction using wavelet - based CLBP for surface defects retrieval of hot rolled steel strip[C]//International Conference on Systems and Control. Sousse，Tunisia：IEEE，2023：849-854.

[20] TAMURA H，MORI S，YAMAWAKI T. Textural features corresponding to visual perception[J]. IEEE Transactions on Systems，Man，and Cybernetics，1978，8(6)：460-473.

[21] LI X，LI J. Application of machine vision in the inspection system of steel plate surface defects[J]. Journal of Physics：Conference Series，2020，1648(4)：42070.

[22] DEVI M R，PRASAD M S G，KARTHIKESWARAN D，et al. Segmentation of surface defects in steel sheet using nearest neighbour region fusion segmentation algorithm[C]//World Conference on Applied Intelligence and Computing. Gwalior，India：IEEE，2024：1085-1089.

[23] HU K，SHI Q，HAN W，et al. On the evolution of temperature and combined stress in a work roll under cyclic thermo-mechanical loadings during hot strip rolling and idling[J]. Materials，2020，13(21)：5054.

[24] HU K，ZHU F，CHEN J，et al. Simulation of thermal stress and fatigue life prediction of high speed steel work roll during hot rolling considering the initial residual stress[J]. Metals，2019，9(9)：966.

[25] SHI S，YU C，ZHOU H，et al. Spatial resolution enhancement with line-scan light-field imaging[J]. Optics Letters，2023，48(20)：5316-5319.

[26] KONG L J，HE Q，LIU G H. Image retrieval based on deep tamura feature descriptor[J]. Multimedia Systems，2024，30(3)：148.

[27] AHMED I T，RASHEED M M，HAMMAD B T. Effective digital image watermark detection technique based on tamura texture features[C]//International Engineering Conference on Sustainable Technology and Development. Erbil，Iraq：IEEE，2023，26-31.

基于纹理特征与集成学习的热轧工作辊表面磨损检测方法研究

Research on Surface Wear Detection Method for Hot Rolling Work Roll Based on Texture Features and Ensemble Learning

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	黄思翰, 陈建鹏, 徐哲, 阎艳, 王国新. 基于大语言模型和机器视觉的智能制造系统人机自主协同作业方法[J]. 机械工程学报, 2025, 61(3): 130-141.
[2]	肖宏, 王阳, 刘秀波, 崔旭浩, 张智海, 金锋. 钢轨波磨检测原理、方法及装置研究进展[J]. 机械工程学报, 2025, 61(10): 191-214.
[3]	罗晨, 邵佳丰, 王伟, 张刚, 周怡君. 考虑镜面反射和阴影的复杂曲面测量视点规划[J]. 机械工程学报, 2024, 60(15): 304-315.
[4]	何赟泽, 李响, 王洪金, 侯岳骏, 张帆, 牟欣颖, 刘浩, 程豪, 李时华, 李杰. 基于可见光和热成像的风机叶片全周期无损检测综述[J]. 机械工程学报, 2023, 59(6): 32-45.
[5]	朱永国, 张文博, 邓正平, 刘春锋. 基于激光跟踪仪和机器视觉的飞机翼身对接装配偏差动态综合修正[J]. 机械工程学报, 2019, 55(24): 187-196.
[6]	刘明周, 蒋倩男, 扈静. 基于面部几何特征及手部运动特征的驾驶员疲劳检测[J]. 机械工程学报, 2019, 55(2): 18-26.
[7]	单忠德, 张飞, 任永新, 张静, 聂军刚. 基于机器视觉铸件布氏硬度在线检测技术研究^*[J]. 机械工程学报, 2017, 53(1): 157-164.
[8]	郝永平, 王永杰, 董福禄, 路超. 平板类微小零件装配控制策略与软件架构研究[J]. 机械工程学报, 2015, 51(4): 193-198.
[9]	王耀南;吴成中;张辉;葛继;周博文;赵科. 医药输液视觉检测机器人关键技术综述[J]. , 2013, 49(7): 130-140.
[10]	张壮雅;张海光;万福平;刘媛媛;胡庆夕. 面向真空注型技术的双组分材料混合脱泡方法[J]. , 2013, 49(23): 28-35.
[11]	刘震;张广军;魏振忠. 基于双平面靶标的多视觉传感器现场全局校准[J]. , 2009, 45(7): 228-232.
[12]	刘震;张广军;魏振忠. 基于一维靶标的多摄像机现场全局校准[J]. , 2009, 45(4): 161-165.
[13]	张宏伟;张国雄;刘书桂;裘祖荣;于复生. 零件轮廓测量机安全监测系统的设计[J]. , 2006, 42(2): 212-215.
[14]	王荣本;储江伟;冯炎;游峰;纪寿文. 一种视觉导航的实用型AGV设计[J]. , 2002, 38(11): 135-138.
[15]	罗振璧;郑力;汪劲松;张伯鹏;潘尚峰;唐晓红. 机器视觉在尺寸在线检测中的应用[J]. , 1994, 30(6): 47-52.