圆柱形锂电池圆周面凹坑缺陷检测方法研究  

Research on Detection Method of Pitting Defects on the Circumferential Surface of Cylindrical Lithium Battery

作　　者：李昱杉 苑玮琦[1,2] LI Yushan;YUAN Weiqi(Computer Vision Group,Shenyang University of Technology,Shenyang 110870,China;Key Laboratory of Machine Vision of Liaoning Province,Shenyang 110870,China)

机构地区：[1]沈阳工业大学视觉检测技术研究所,沈阳110870 [2]辽宁省机器视觉重点实验室,沈阳110870

出　　处：《微处理机》2023年第5期57-60,共4页Microprocessors

摘　　要：针对常规圆柱形锂电池出厂检验中对圆周面上的凹坑缺陷特征识别度较低的问题,基于对识别模型的灰度分与光照分布的详细分析,提出一种光照分布修正方法。方法通过线性变换将像素点灰度值处理到同一基准线,以去除光照对缺陷检测的影响;通过增强缺陷与背景区域的对比度,基于灰度分布曲线过渡线段提取缺陷特征,并在实验中进行关键参数选择完成测试。测试结果表明该方法增强了凹坑缺陷检测的准确性、适用性和实时性,有效地促进了电池产品的出厂安全性。In order to solve the problem of low recognition degree of pitting defect features on the circumferential surface in the factory inspection of conventional cylindrical lithium batteries,based on the detailed analysis of gray scale and illumination distribution of the recognition model,a correction method of illumination distribution is proposed.In the method,the gray values of pixels are processed to the same baseline by linear transformation to remove the influence of illumination on defect detection.By enhancing the contrast between defects and background areas,defect features are extracted based on the transition line segment of gray distribution curve,and key parameters are selected in the experiment to complete the test.The test results show that the method enhances the accuracy,applicability and real-time performance of pitting defect detection,and effectively promotes the factory safety of battery products.

关 键 词：视觉检测 锂电池凹坑缺陷 光照分布修正 过渡线段 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]