基于机器视觉的冷轧带材跑偏量智能检测方法  

An intelligent detection method for cold-rolled strip deviation based on machine vision

作　　者：段伯伟 王东城 徐扬欢 邢佳文 刘宏民 Duan Bowei;Wang Dongcheng;Xu Yanghuan;Xing Jiawen;Liu Hongmin(National Engineering Research Center for Equipment and Technology of Cold Strip Rolling,Yanshan University,Qinhuangdao 066004,China;National Key Laboratory of Metal Forming Technology and Heavy Equipment,China National Heavy Machinery Research Institute Co.,Ltd,Xi’an 710032,China)

机构地区：[1]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 [2]中国重型机械研究院股份公司金属成形技术与重型装备全国重点实验室,西安710032

出　　处：《仪器仪表学报》2024年第9期55-66,共12页Chinese Journal of Scientific Instrument

基　　金：国家自然科学基金项目(U21A20118);河北省自然科学基金面上项目(E2023203065);金属成形技术与重型装备全国重点实验室(中国重型院)开放课题项目(S2208100.W04)资助。

摘　　要：针对冷轧生产过程中缺少实测跑偏量引起的板形控制精度低、断带等问题,提出一种基于机器视觉的冷轧带材跑偏量智能检测方法。以经典UNet网络为基础搭建一种轻量化的网络结构,用于冷轧带材智能分割。采用MobileNetV2替换UNet原始收缩路径并在连接结构中嵌入通道注意力ECA_Module,在有效降低网络参数量的同时突出对目标特征的感知能力。采用此网络训练得到带材区域分割模型(SRS_M),其精度指标交并比平均值(mIoU)和像素精度平均值(mPA)分别达到了98.83%和99.36%,单张图像推理时间为40.57 ms。以SRS_M模型为基础,结合边缘位置提取算法,建立带材跑偏量检测模型。通过现场安装的边缘检测装置采集1503个跑偏量样本数据,对方法进行验证。其中,92.82%样本的绝对误差在±2 mm,全部样本绝对误差均在±3.5 mm,证明了方法的有效性。The lack of measured deviation of the strip during cold-rolled process may cause problems of reduction in flatness the control accuracy and strip breakage.An intelligent detection method for cold-rolled strip deviation based on machine vision is proposed in this article.A lightweight network structure is constructed,which is based on classic network UNet for intelligent segmentation of cold-rolled strips.The MobileNetV2 is used to replace the original contraction path of UNet and the channel attention ECA_Module is embedded in the connection structure,which effectively reduces the amount of network parameters while enhancing the perception ability of target features.The strip region segmentation model of the cold-rolled strip(SRS_M)can be obtained by training this network.The accuracy indicators mIoU and mPA of SRS_M could reach 98.83%and 99.36%,respectively.The running time of a single image is 40.57 ms.An intelligent detection model of strip deviation is formulated by combining SRS_M and edge position extraction algorithm.The 1503 deviation sample data are collected through an edge detection device installed on site,which is used to evaluate the proposed method.The results show that the absolute error of 92.82%of the samples is within the range of±2 mm,and the absolute error of all samples is within the range of±3.5 mm,which shows the effectiveness of the proposed method in this article.The lack of measured deviation of strip during cold-rolled process will cause problem like reduction in flatness control accuracy and strip breakage.An intelligent detection method for cold-rolled strip deviation based on machine vision is proposed in this paper.A lightweight network structure is built based on classic network UNet for intelligent segmentation of cold-rolled strip.The MobileNetV2 is used to replace the original contraction path of UNet and the channel attention ECA_Module is embedded in the connection structure,which effectively reduces the amount of network parameters while enhancing the perception ability of

关 键 词：冷轧带材 跑偏量 机器视觉 智能语义分割 通道注意力机制 

分 类 号：TH89[机械工程—仪器科学与技术] TP391[机械工程—精密仪器及机械]