Research on Image Flatness Detection Method Based on Deep Learning

doi:10.3901/JME.260185

Abstract

Abstract: Intelligent manufacturing is recognized as a key direction for the transformation and upgrading of China’s steel industry. The accurate digital perception of process parameters and product quality is regarded as its foundation and priority. Flatness is a critical parameter for cold-rolled strip products, and the development and application of online flatness detection technology is thus essential for the production of high-quality strip. However, the deployment rate of flatness meters in the domestic steel industry remains low, with imported equipment being prohibitively expensive. Consequently, most enterprises still rely on manual assessment, leading to inconsistent product quality and insufficient levels of intelligentization. There is an urgent need for innovative research into flatness detection mechanisms, with the goal of reducing equipment costs and enhancing the universality and adaptability of detection technologies without compromising detection accuracy. In recent years, the theory and application of machine vision, particularly deep learning, have developed rapidly. Applying machine vision technology to flatness detection is not only theoretically feasible but also an inevitable trend in the advancement of intelligent manufacturing. Through theoretical and technological innovations, a series of multi-level intelligent models were developed to accomplish strip position detection, deep representation of image and strain flatness, quantitative mapping of representation factors, and finally, a precise image-to-strain flatness mapping. The proposed models and framework provide insights and a concrete scheme for intelligent flatness detection, paving the way for the industry’s technological transformation.

Key words: cold rolled strip, image flatness, strain flatness, deep learning, machine vision, intelligent detection

CLC Number:

TG335

XU Yanghuan, WANG Dongcheng, LIU Hongmin, DANG Liying, LIU Yunfei, YANG Aimin. Research on Image Flatness Detection Method Based on Deep Learning[J]. Journal of Mechanical Engineering, 2025, 62(6): 185-196.

Add to citation manager EndNote|Reference Manager|ProCite|BibTeX|RefWorks

URL: http://www.cjmenet.com.cn/EN/10.3901/JME.260185

http://www.cjmenet.com.cn/EN/Y2025/V62/I6/185

References

[1] 宁楠，惠宁. 人工智能驱动制造业新质生产力发展——基于全要素生产率的视角[J]. 经济问题探索，2025(1)：1-22.

NING Nan，HUI Ning. Artificial intelligence driving the development of new quality productive forces in manufacturing industry：Based on the total factor productivity perspective[J]. Inquiry into Economic Issues，2025(1)：1-22.

[2] 李其伦. 中国制造业数字化与绿色化融合协同发展测度及演化分析[J]. 经济问题探索，2024(7)：119-132.

LI Qilun. Measurement and evolution analysis of the integration and coordinated development of digitalization and greenization in China’s manufacturing industry[J]. Inquiry into Economic Issues，2024(7)：119-132.

[3] 郭克莎，彭继宗. 制造业在中国新发展阶段的战略地位和作用[J]. 中国社会科学，2021(5)：128-149.

GUO Kesha，PENG Jizong. The manufacturing sector’s strategic position and role in China’s new stage of development[J]. Social Sciences in China，2021(5)：128-149.

[4] 王国栋，张殿华，孙杰. 建设数据驱动的钢铁材料创新基础设施加速钢铁行业的数字化转型[J]. 冶金自动化，2023，47(1)：2-9.

WANG Guodong，ZHANG Dianhua，SUN Jie. Creating data-driven innovation infrastructure for iron and steel material to accelerate digital transformation of iron and steel industry[J]. Metallurgical Industry Automation，2023，47(1)：2-9.

[5] 殷瑞钰. 新世纪以来中国炼钢-连铸的进步及命题[J]. 中国冶金，2014，24(8)：1-9.

YIN Ruiyu. Progress and propositions of Chinese steelmaking and continuous casting since the new century[J]. China Metallurgy，2014，24(8)：1-9.

[6] 黄庆学. 高品质钢铁板带轧制关键装备与技术研究进展[J]. 机械工程学报，2023，59(20)：34-63.

HUANG Qingxue. Research progress on key equipment and technology of high quality steel plate and strip rolling[J]. Journal of Mechanical Engineering，2023，59(20)：34-63.

[7] 何安瑞，荆丰伟，刘超. 薄板坯连铸连轧过程控制技术的发展、应用及展望[J]. 轧钢，2020，37(3)：1-7.

HE Anrui，JING Fengwei，LIU Chao. Development，application and prospect of process control technologies of thin slab continuous casting and rolling[J]. Steel Rolling，2020，37(3)：1-7.

[8] 梁青艳，孙彦广，李文兵，等. 基于多智能体技术的钢铁生产流程仿真及应用[J]. 冶金自动化，2024，48(3)：92-100.

LIANG Qingyan，SUN Yanguang，LI Wenbing，et al. Simulation and application of steel production process basedon multi-agent technology[J]. Metallurgical Industry Automation，2024，48(3)：92-100.

[9] 何安瑞，刘超，邵健. 热轧带钢高效与智能轧制新技术研究进展及应用[J]. 轧钢，2024，41(5)：38-50.

HE Anrui，LIU Chao，SHAO Jian. Research progress and application of new advanced and intelligent rolling technologies for hot rolled strip[J]. Steel Rolling，2024，41(5)：38-50.

[10] 张殿华，丁成砚，王云龙，等. 板带轧制数字化技术进步与发展趋势[J]. 轧钢，2024，41(5)：51-65.

ZHANG Dianhuan，DING Chengyan，WANG Yunlong，et al. Developments and prospects of digital technique in strip rolling process[J]. Steel Rolling，2024，41(5)：51-65.

[11] 孙杰，陈树宗，王云龙，等. 冷连轧关键质量指标与轧制稳定性智能优化控制技术[J]. 钢铁研究学报，2022，34(12)：1387-1397.

SUN Jie，CHEN Shuzong，WANG Yunlong，et al. Intelligent optimization control technology for key quality indexes and rolling stability of tandem cold rolling[J]. Journal of Iron and Steel Research，2022，34(12)：1387-1397.

[12] 刘晔，闫博，刘长鑫. 基于厚板生产大数据的智能监控系统研究[J]. 控制工程，2020，27(12)：2204-2209. LIU Ye，YAN Bo，LIU Changxin. Research on intelligent monitoring system based on big data of heavy plate production[J]. Control Engineering of China，2020，27(12)：2204-2209.

[13] 孙建亮，孙孟乾，郭贺松，等. 热轧带材多缺陷和单缺陷表面质量综合预报[J]. 钢铁，2021，56(1)：51-58. SUN Jianliang，SUN Mengqian，GUO Hesong，et al. Comprehensive prediction of surface quality of hot-rolled strip with multiple and single defects[J]. Iron and Steel，2021，56(1)：51-58.

[14] 王东城，徐扬欢，段伯伟，等. 数据驱动的热轧带钢边部线状缺陷智能预报模型[J]. 钢铁，2020，55(11)：82-90.

WANG Dongcheng，XU Yanghuan，DUAN Bowei，et al. Data-driven intelligent prediction model of edge seam defects for hot rolling strip[J]. Iron and Steel，2020，55(11)：82-90.

[15] 王晓雯，张勇军，郭强，等. 基于轧制机理和混合神经网络的热轧精轧带宽预测[J]. 中国冶金，2023，33(2)：114-120.

WANG Xiaowen，ZHANG Yongjun，GUO Qiang，et al. Prediction of hot strip width based on rolling mechanism and hybrid neural network[J]. China Metallurgy，2023，33(2)：114-120.

[16] 刘阳，郜志英，周晓敏，等. 工业数据驱动下薄板冷轧颤振的LSTM智能预报[J]. 机械工程学报，2020，56(11)：121-131.

LIU Yang，GAO Zhiying，ZHOU Xiaoming，et al. Industrial data-driven intelligent forecast for chatter of cold rolling of thin strip with LSTM recurrent neural network[J]. Journal of Mechanical Engineering，2020，56(11)：121-131.

[17] LIU Yi，ZHANG Changsheng，DONG Xingjun. A survey of real-time surface defect inspection methods based on deep learning[J]. Artificial Intelligence Review，2023，56(10)：12131-12170.

[18] 王辉，徐佳文，严如强. 基于多尺度注意力深度强化学习网络的行星齿轮箱智能诊断方法[J]. 机械工程学报，2022，58(11)：133-142.

WANG Hui，XU Jiawen，YAN Ruqiang. Multi-scale attention based deep reinforcement learning for intelligent fault diagnosis of planetary gearbox[J]. Journal of Mechanical Engineering，2022，58(11)：133-142.

[19] 刘振宇，吴思炜，丛津功，等. 热轧钢材组织演变和力学性能预测—半经验模型、机器学习到工业大模型[J]. 轧钢，2024，41(5)：24-37，50.

LIU Zhenyu，WU Siwei，CONG Jingong，et al. Development of the technology to predict microstructurel evolution and changes of mechanical properties for hot rolled steels—the path from semi-empirical models through machine learning to industrial foundation models[J]. Steel Rolling，2024，41(5)：24-37，50.

[20] 雷亚国，李熹伟，李响，等. 面向机械设备通用健康管理的智能运维大模型[J]. 机械工程学报，2025，61(6)：1-13.

LEI Yaguo，LI Xiwei，LI Xiang，et al. Research on large model for general prognostics and health management of machinery[J]. Journal of Mechanical Engineering，2025，61(6)：1-13.

[21] 白辰甲，许华哲，李学龙. 大模型驱动的具身智能：发展与挑战[J]. 中国科学：信息科学，2024，54(9)：2035-2082.

BAI Chenjia，XU Huazhe，LI Xuelong. Embodied-AI with large models：Research and challenges[J]. Scientia Sinica Informationis，2024，54(9)：2035-2082.

[22] 刘文仲. 中国轧钢自动化现状及实现轧钢智能化的思考[J]. 冶金自动化，2016，40(6)：1-5.

LIU Wenzhong. Status of rolling automation in China and thinking about realization of intelligent rolling[J]. Metallurgical Industry Automation，2016，40(6)：1-5.

[23] 彭艳. 冶金轧制设备技术数字化智能化发展概述[J]. 燕山大学学报，2020，44(3)：218-237.

PENG Yan. Review on development of digital and intelligent metallurgical rolling equipment technology[J]. Journal of Yanshan University，2020，44(3)：218-237.

[24] 华长春，陈树宗，李旭，等. 轧制全流程质量智能建模、故障诊断与协同稳健控制的研究现状与展望[J]. 冶金自动化，2022，46(6)：38-47.

HUA Changchun，CHEN Shuzong，LI Xu，et al. Research status and prospect of intelligent modeling fault diagnosis and cooperative robust control for whole rolling process quality[J]. Metallurgical Industry Automation，2022，46(6)：38-47.

[25] 孙文权，袁铁衡. 我国冷轧生产技术的发展探讨[J]. 冶金动力，2022(2)：1-6.

SUN Wenquan，YUAN Tieheng. Discussion on the development of cold rolling production technology in China[J]. Metallurgical Power，2022(2)：1-6.

[26] 孙杰，叶俊成，彭文，等. 板带材生产全流程数字孪生模型及信息物理系统研究现状与趋势[J]. 轧钢，2022，39(6)：37-45.

SUN Jie，YE Juncheng，PENG Wen，et al. Research status and tendency of digital twin model and cyber physical system for sheet and strip production process[J]. Steel Rolling，2022，39(6)：37-45.

[27] 康永林，朱国明，陈贵江，等. 中国薄宽带钢无头轧制技术最新进展[J]. 钢铁，2023，58(7)：1-8.

KANG Yonglin，ZHU Guoming，CHEN Guijiang，et al. Latest development of wide thin strip endless rolling technology in China[J]. Iron and Steel，2023，58(7)：1-8.

[28] 刘宏民，于华鑫，王东城，等. 冷轧带钢板形测控技术的发展状况和关键问题[J]. 钢铁，2022，57(11)：22-32. LIU Hongmin，YU Huaxin，WANG Dongcheng，et al. Development state and key problems on flatness measurement and control technology of cold strip steel rolling[J]. Iron and Steel，2022，57(11)：22-32.

[29] 罗超，刘晓，任忠凯，等. 金属复合极薄带制备工艺研究进展[J]. 材料导报，2023，37(12)：201-206.

LU Chao，LIU Xiao，REN Zhongkai，et al. Innovative strategy on frontier flatness measurement and control technology of cold strip steel rolling[J]. Materials Reports，2023，37(12)：201-206.

[30] 刘宏民，于华鑫，王东城，等. 冷轧带钢板形测控前沿技术的创新方略[J]. 钢铁，2023，58(4)：77-86.

LIU Hongmin，YU Huaxin，WANG Dongcheng，et al. Innovative strategy on frontier flatness measurement and control technology of cold strip steel rolling[J]. Iron and Steel，2023，58(4)：77-86.

[31] 杨利坡，于华鑫，张永顺，等. 冷轧带钢虚拟仪器板形测控系统及其应用[J]. 机械工程学报，2018，54(14)：1-7.

YANG Lipo，YU Huaxin，ZHANG Yongshun，et al. Shape detection and control system of cold rolling strip based on the virtual instrument and its industrial application[J]. Journal of Mechanical Engineering，2018，54(14)：1-7.

[32] WANG Dongcheng，LIU Hongmin，LIU Jun. Research and development trend of shape control for cold rolling strip[J]. Chinese Journal of Mechanical Engineering，2017，30(5)：1248-1261.

[33] 刘宏民，刘军，于丙强，等. 带钢冷轧机整辊无线式板形仪和智能板形控制系统[J]. 机械工程学报，2017，53(12)：87-93.

LIU Hongmin，LIU Jun，YU Bingqiang，et al. Entire roller wireless shape meter and intelligent shape control system for cold steel strip mills[J]. Journal of Mechanical Engineering，2017，53(12)：87-93.

[34] 刘宏民，于丙强，杨利坡，等. 整辊镶块智能型冷轧带钢板形仪的研制与工业应用[J]. 钢铁，2011，46(12)：86-89.

LIU Hongmin，YU Bingqiang，YANG Lipo，et al. Development of cold strip shape meter with entire roller inlayed block intelligence and its industrial application[J]. Iron and Steel，2011，46(12)：86-89.

[35] 宿磊，王立建，祁阳，等. 基于IADSA深度迁移网络的金属表面缺陷检测[J]. 机械工程学报，2023，59(24)：46-55.

SUN Lei，WANG Lijian，QI Yang，et al. Metal surface defect detection based on IADSA deep transfer network[J]. Journal of Mechanical Engineering，2023，59(24)：46-55.

[36] SONG Kechen，CUI Wenqi，YU Han，et al. SAM Era：Can it segment any industrial surface defects[J]. Cmc- computers Materials & Continua，2024，78(3)：3953-3969.

[37] DUAN Bowei，WANG Dongcheng，MA Yuhua，et al. Multisource data-driven intelligent method for detecting surface defects in cold-rolled copper strips[J]. Engineering Applications of Artificial Intelligence，2025，152：110730.

[38] XU Yanghuan，WANG Dongcheng，LIU Hongmin，et al. Flatness defect recognition method of cold rolling strip with a new stacked generative adversarial network[J]. Steel Research International，2022，93(11)：2200284.

[39] XU Yanghuan，WANG Dongcheng，LIU Hongmin，et al. Intelligent representation method of image flatness for cold rolled strip[J]. Journal of Iron and Steel Research International，2024，31(5)：1177-1195.

[40] XU Yanghuan，WANG Dongcheng，DUAN Bowei，et al. Data-driven flatness intelligent representation method of cold rolled strip[J]. Journal of Iron and Steel Research International，2023，30(5)：994-1012.