摘要
针对人工肉眼检测高频变压器点胶质量存在劳动强度大、检测效率不高及容易错检、漏检等问题,采用机器视觉方法,设计图像处理算法提取点胶特征进行检测。分别采用HSV颜色空间法和模板匹配法提取特征,详细描述了图像处理分析的步骤,并进行了实验验证,实验结果表明两种检测方法均能快速准确地测出高频变压器的点胶质量,极大地提高了检测效率和准确率。
At present,the dispensing quality check of high frequency transformer is performed by eyes examination.It always suffer the problems of great labor intensity and low production effciency,moreover,it may cause wrong detection and leak inspection.The machine vision inspection is adopted and the image processing algorithm is designed.HSV color space method and template matching method were used to extract dispensing features respectively.The steps of image processing analysis were described in detail,and experiments were conducted to investigate the accuracy.The experimental results show that both methods have a good performance on detecting the dispensing quality,and the detection effciency as well as detection accuracy were greatly improved.
作者
代海东
向芷漫
林沁庭
邹祥和
DAI Haidong;XIANG Zhiman;LIN Qinting;ZOU Xianghe(College of Electrical Engineering and Automation,Jiangxi University of Science and Technology,Ganzhou 341000,China)
出处
《电气应用》
2019年第S01期144-147,共4页
Electrotechnical Application
基金
2018年江西理工大学校级大学生创新创业训练计划项目(DC2018-041)
关键词
机器视觉
高频变压器
点胶
machine vision
high frequency transformer
dispensing
