天津科技大学 机械工程学院 天津市轻工与食品机械装备集成设计与在线监控重点实验室,天津 300457
程颖(1982— ),女,副教授,博士,从事光学自由曲面设计、机器视觉检测方面的研究。chengyingcd@163.com
秦浩森(2000— ),男,硕士研究生,从事机器视觉检测方面的研究。15598750997@163.com
收稿:2025-10-22,
修回:2025-12-26,
纸质出版:2026-03-20
移动端阅览
程颖,秦浩森,毕德学等.基于动态视觉标定的大尺寸钢板平面度在线检测方法[J].光学技术,2026,52(02):230-235.
CHENG Ying,QIN Haosen,BI Dexue,et al.An online detection method for the flatness of large-sized steel plates based on dynamic visual calibration[J].OPTICAL TECHNIQUE,2026,52(02):230-235.
为了消除大尺寸钢板平面度动态高精度测量中,滚筒传输速度非均匀波动引发的扫描畸变及长期负载下滚筒引入的系统误差,构建了基于双目线激光与非编码视觉标记的运动补偿检测系统。核心创新是在钢板表面布设非编码视觉标记点,集成亚像素级角点追踪与速度矢量实时解算算法,实现钢板运动速度高精度动态标定,消除线扫描对机械传输速度稳定性的依赖。实验表明,系统在5Hz的采样频率下
平面度检测误差稳定在0.2mm以内;通过滚筒磨损仿真模型与实验验证,滚筒形变达6mm的极端工况,测量误差仍≤0.2mm,鲁棒性良好。三维点云处理中,最小二乘法拟合基准平面较主成分分析法,可将最终测量误差减低20%。方法为工业现场大尺寸板材动态、高精度、高鲁棒性在线检测提供了可靠的解决方案。
To eliminate the scanning distortion caused by the non-uniform fluctuation of the roller transmission speed and the system error introduced by the rollers under long-term load in the dynamic high-precision measurement of the flatness of large-sized steel plates
a motion compensation detection system based on binocular line laser and non-coded visual markers was constructed. The core innovation lies in the deployment of non-coded visual marker points on the surface of the steel plate
integrating sub-pixel corner tracking and real-time speed vector calculation algorithms to achieve high-precision dynamic calibration of the steel plate's movement speed
thereby eliminating the dependence of line scanning on the stability of mechanical transmission speed. Experiments show that the system's flatness detection error is stable within 0.2mm at a sampling frequency of 5Hz. Through the simulation model of roller wear and experimental verification
even in the extreme condition where the roller deformation reaches 6mm
the measurement error is still ≤0.2mm
demonstrating good robustness. In the processing of three-dimensional point clouds
the least squares method for fitting the reference plane can reduce the final measurement error by 20% compared to the principal component analysis method. This method provides a reliable solution for the dynamic
high-precision
and highly robust online detection of large-sized plates in industrial fields.
杨璐雅 . 钢板表面缺陷图像检测与分类识别技术研究 [D ] . 西安 : 西安电子科技大学 , 2023 .
Yang Luyi . Research on image detection and classification recognition technology for surface defects of steel plates [D ] . Xi'an University of Electronic Science and Technology , 2023
王韬 . 基于线结构光的钢板轮廓检测系统研究 [D ] . 太原 : 太原科技大学 , 2024
Wang Tao . Research on steel plate profile detection system based on linear structured light [D ] . Taiyuan University of Science and Technology , 2024
Liu Y , Ou P , Xu X , et al . Multi-line structured light binocular vision stereo matching method via coarse-to-fine spatial geometric constraints [J ] . Optics and Laser Technology , 2024 , 176 : 110950 .
时浩 , 邓高旭 , 李正楠 , 等 . 基于激光点云的中厚板表面平整度测量 [J ] . 光学精密工程 , 2024 , 32 ( 16 ): 2464 — 2473 .
Shi Hao , Deng Gaoxu , Li Zhengnan , et al . Measurement of surface flatness of medium and thick plates based on laser point cloud [J ] . Optics and Precision Engineering , 2024 , 32 ( 16 ): 2464 — 2473 .
杨璐雅 . 钢板表面缺陷图像检测与分类识别技术研究 [D ] . 西安 : 西安电子科技大学 , 2023 .
Yang Luya . Research on image detection and classification recognition technology of surface defects on steel plates [D ] . Xidian University , 2023 .
Wang L , Yang H , Jiang J , et al . A cone array calibration-based line laser triangulation method for freeform surface measurement independent of motion device precision [J ] . Optics and Lasers in Engineering , 2025 , 195109272 — 109272 .
李政锋 . 基于双目视觉和线结构光的工件三维重建与测量 [D ] . 厦门 : 厦门大学 , 2021 .
Li Zhengfeng . Three-dimensional reconstruction and measurement of workpieces based on binocular vision and linear structured light [D ] . Xiamen University , 2021 .
Chen Y , Jiang W , Luo Z , et al . A novel 3D reconstruction method with a binocular-line laser system [J ] . Measurement , 2024 , 227 : 114238 .
Fu Y , Liu B , Zhong K , et al . Virtual binocular stereo vision measurement method without phase unwrapping [J ] . Optics Communications , 2025 , 585 : 131869 — 131869 .
Cao M , Zhang G , Chen Y . Stereo matching of light-spot image points in light pen in binocular stereo visual [J ] . Optik-International Journal for Light and Electron Optics , 2014 , 125 ( 3 ): 1366 — 1370 .
刘盼 , 王金海 . 基于平行双目立体视觉的测距系统 [J ] . 计算机应用 , 2012 , 32 ( S2 ): 162 — 164 .
Liu Pan , Wang Jinhai . Distance measurement system based on parallel binocular stereo vision [J ] . Computer Applications , 2012 , 32 ( S2 ): 162 — 164 .
万聪灵 , 余学才 , 蒋波 , 等 . 线结构光双目传感器钢坯轮廓测量标定方法 [J ] . 计量学报 , 2014 , 35 ( 01 ): 34 — 38 .
Wan Congling , Yu Xuecai , Jiang Bo , et al . Calibration method for billet profile measurement with line-structured light binocular sensor [J ] . Acta Metrologica Sinica , 2014 , 35 ( 01 ): 34 — 38 .
曹淞翔 . 基于双目视觉的三维重建 [J ] . 通讯世界 , 2018 , 25 ( 12 ): 236 — 238 .
Cao Songxiang . 3D reconstruction based on binocular vision [J ] . Communications World , 2018 , 25 ( 12 ): 236 — 238 .
张宇 , 李毅红 , 陈平 , 等 . 基于机器视觉的带钢板边浪平直度测量方法 [J ] . 中国测试 , 2024 , 50 ( 09 ): 91 — 97 .
Zhang Yu , Li Yihong , Chen Ping , et al . Measurement method of edge wave straightness of steel plate based on machine vision [J ] . Chinese Journal of Testing , 2024 , 50 ( 09 ): 91 — 97 .
郑游 . 基于手持式的双目扫描三维重建方法研究 [D ] . 武汉 : 武汉工程大学 , 2023 .
Zheng You . Research on 3D reconstruction method based on handheld binocular scanning [D ] . Wuhan Institute of Technology , 2023 .
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621