山东农业大学学报（自然科学版）2025，Vol.56Issue(4)：681-689,9.DOI:10.3969/j.issn.1000-2324.2025.04.014
基于机器视觉的压力表智能读数系统
Intelligent Reading System for Pressure Gauges Based on Machine Vision
摘要
Abstract
Pressure gauges are essential measurement tools in industrial production,capable of effective measurement of mechanical working pressure.Their calibration process often requires separate equipment,such as pressure sources and controllers,which relies on wired connections.This setup is prone to connector wear and cable clutter.Thus,we develop an all-in-one intelligent reading calibration system of pressure gauges to achieve the organic fusion of equipment and cables.It includes a fully automatic air pressure generator,a visual inspection device,a quick-change lifting device,and a pressure gauge reading method based on the improved Mask R-CNN deep learning model.The test results of the pneumatic pressure generator show that the absolute error is within 0.002,the control stability and cycle stability are in accordance with the standards of the calibration regulations,and the pressure generation reaches 600 kPa,meeting performance design requirements.For pressure gauge reading test,the relative error is within 1.05%for small-range gauges and 4.57%for larger-range gauges,with a higher identification accuracy.关键词
压力检定/视觉检测/全自动气压发生器/Mask R-CNNKey words
Pressure calibration/visual detection/fully automatic air pressure generator/Mask R-CNN分类
信息技术与安全科学引用本文复制引用
王必军,侯家彬,张开兴,郑来臣,周长安,刘鑫..基于机器视觉的压力表智能读数系统[J].山东农业大学学报（自然科学版）,2025,56(4):681-689,9.基金项目
山东省科技型中小企业创新能力提升工程项目(2022TSGC2454) （2022TSGC2454）
山东省自然科学基金青年项目(ZR2021QE238) （ZR2021QE238）
国家自然科学基金(52305543) （52305543）
