摘要
针对实际生产制造中纽扣电池的划痕、凹坑或污渍等缺陷自动检测问题,提出一种基于机器视觉的纽扣电池缺陷检测方法。利用同轴光源和多角度条形组合光源,搭建了视觉成像系统,获取高质量缺陷图像。然后对图像进行预处理,使用简化的组合缺陷检测算法进行检测,形成了一套完整、高效的纽扣电池缺陷检测系统。最后对纽扣电池的划痕、凹坑及污渍缺陷进行批量实验,结果表明:三类缺陷的检测成功率均超过98%,验证了该检测系统的有效性。
It proposes a method of using machine vision for detecting button battery defects such as scratches,pits and rust spots.An imaging system has been developed by using a coaxial light source and a multi-angle combination bar light source to obtain high quality defect images. After the image is preprocessed,the system uses a simplified and combined defect detection algorithm to detect the button battery defects efficiently. Experiments have been carried out on some button batteries with scratches,pits and stains. The results show that the detection success rate of the three types of defects was over 98%. It shows that the detection system is promising.
作者
梁经伦
叶国良
郭建文
吴佳鑫
LIANG Jing-lun;YE Guo-liang;GUO Jian-wen;WU Jia-xin(School of Mechanical Engineering,Dongguan University of Technology,Guangdong Dongguan 523808,China)
出处
《机械设计与制造》
北大核心
2023年第3期225-229,共5页
Machinery Design & Manufacture
基金
广东省普通高校重点领域专项(2020ZDZX2016、2021ZDZX1067)
2019年东莞市社会科技发展(重点)项目(2019507140209)
2020年东莞市科技特派员项目(20201800500282、20201800500212)
2017年广东省普通高校省级重大科研项目(2017KZDXM082)。
关键词
机器视觉
纽扣电池
表面缺陷检测
Machine Vision
Button Battery
Surface Defect Detection
