摘要
随着5G与物联网技术的发展,工厂的作业系统与科研院校的实验室可以整合为一个整体,这就使得实时,可更新的缺陷识别算法的研发和应用成为可能。对比研究了四个表面缺陷数据集与对应的检测方法,并在此基础上提出了若想要达到工业生产级别的应用,需要解决的关键技术问题。
With the development of 5 G and Internet of things technology,the factory operating system and the laboratory of scientific research institutions can be integrated into a whole,which makes the development and application of real-time,renewable defect identification algorithm possible.Four data sets of surface defects and corresponding detection methods are compared and studied in this paper.On this basis,the key technical problems that need to be solved in order to achieve industrial application level are proposed.
作者
肖克来提
Shockletti(Xinjiang Normal University,Xinjiang 830054,China)
出处
《电子技术（上海）》
2020年第8期189-191,194,共4页
Electronic Technology
基金
新疆师范大学优秀青年教师科研启动基金特别资助(XJNU201508)
关键词
控制技术
缺陷检测
深度学习
表面缺陷
数据处理
control technology
defect detection
deep learning
surface defect
data processing
