锂电池极片轧机轧辊表面缺陷检测与识别  

Detection and recognition of roll surface defects in lithium battery pole rolling mill

作　　者：肖艳军[1] 齐浩 周围[1] 彭凯[1] 孟召宗 张雪辉 Xiao Yanjun;Qi Hao;Zhou Wei;Peng Kai;Meng Zhaozong;Zhang Xuehui(School of Mechanical Engineering,Hebei University of Technology,Tianjin 300000,China)

机构地区：[1]河北工业大学机械工程学院

出　　处：《电子测量与仪器学报》2019年第10期148-156,共9页Journal of Electronic Measurement and Instrumentation

基　　金：河北省重点研发计划(18214407D);河北省科技型中小企业创新英才项目(179A7631H);河北省科技项目(16211927)资助

摘　　要：为了能准确的对锂电池极片轧机轧辐表面的缺陷类型做出检测,提出了一种基于图像处理的轧辐表面缺陷检测与识别方法。首先,针对轧辐表面缺陷进行了类别划分,并设计了相应的轧辐表面质量检测系统,实现图像的收集与提取;然后针对收集的轧辐表面图片,利用前景提取和图像增强实现初步的预处理;最后利用特征提取与特征归类的方法对经过预处理的图像进行缺陷归类。通过对40组轧辐缺陷样本进行试验测试,结果表明,轧辐表面缺陷检测的准确度达到90%,证明了该方法的精度和有效性,该方法能够有效鉴别轧辐表面的缺陷类型,对锂电池极片轧辐的维护有着指导作用。In order to accurately detect the defect types on the roll surface of the lithium battery pole piece mill,this paper proposes a method based on image processing for the detection and identification of roll surface defects.Firstly,the classification of roll surface defects is carried out,and the corresponding roll surface quality inspection system is designed to realize image collection and extraction.Then,for the collected roll surface image,preliminary preprocessing is realized by foreground extraction and image enhancement,and then the defect classification is performed on the preprocessed image by feature extraction and feature classification.The test results of 40 sets of roll defect samples show that the accuracy of roll surface defect detection reaches 90%,which proves the accuracy and effectiveness of the method.The method can effectively identify the type of defects on the surface of the roll and has a guiding role in the maintenance of the pole piece of the lithium battery.

关 键 词：极片轧机轧辐 前景提取 缺陷提取 缺陷归类 

分 类 号：TP751.2[自动化与计算机技术—检测技术与自动化装置]