基于深度学习的函数发生器远程校准系统  

A remote calibration system for function generators based on deep learning

作　　者：曹志清 章乐 梅国健 CAO Zhiqing;ZHANG Le;MEI Guojian(College of Information Engineering,China Jiliang University,Hangzhou 310018,China;Zhejiang Institute of Metrology,Hangzhou 310018,China)

机构地区：[1]中国计量大学信息工程学院,浙江杭州310018 [2]浙江省计量科学研究院,浙江杭州310018

出　　处：《中国计量大学学报》2023年第2期193-200,共8页Journal of China University of Metrology

基　　金：国家自然科学基金项目(No.62175223)。

摘　　要：目的:针对不具备通信接口电子测量仪器的远程自动化校准问题,设计了一种基于校准流程组态和深度学习的远程校准系统。方法:系统服务端负责依据校准任务需求进行远程组态调度。本地用户端通过解析下派的指令文件和指导文件依次执行测量步骤,结合OCR深度学习技术对摄像头捕捉的仪器面板画面进行示值读数识别,并在DBNet和CRNN网络的基础上对算法进行场景适应性改进,在校准结束时自动生成校准结果报告。结果:实现了无法进行程控通信的电子测量仪器自动化校准过程,并且在仪器倾斜和示值小数点模糊等情况下仍有较高的识别准确率。结论:本文通过结合OCR深度学习技术的远程自动系统可以有效提高现场校准的效率,在电子测量仪器校准工作中具有一定的实用价值。Aims:Aiming at the problem of remote automatic calibration of electrical measurement instruments without communication interface,a remote calibration system based on calibration process configuration and deep learning was designed.Methods:The system server was responsible for remote configuration and scheduling based on the calibration task requirements.The local user side performed the measurement steps sequentially by parsing the assigned command and instruction files.It combined the OCR deep learning technology to identify the readings of the instrument panel captured by the camera and adapted the algorithm to the scenario based on DBNet and CRNN networks to automatically generate a calibration result report at the end of the calibration.Results:The system realized the automatic calibration of the electronic measurement instrument without programmable communication and provided high recognition accuracy in the situation of tilted and blur display.Conclusions:In this paper,a remote automation system combined with the OCR deep learning technology can effectively improve the efficiency of on-site calibration and has a practical value in the calibration of electronic measurement instruments.

关 键 词：深度学习 远程校准 组态调度 OCR技术 

分 类 号：TB97[一般工业技术—计量学]