自动化与信息工程2025，Vol.46Issue(6)：17-23,7.DOI:10.12475/aie.20250603
基于机器视觉的毫升级容量瓶注水定容方法
Machine Vision-based Water Injection and Constant Volume Method for Milliliter-scale Volumetric Flasks
摘要
Abstract
To address the issue of significant drip errors during the automatic machine vision-based water injection and constant volume process for milliliter-scale volumetric flasks,four imaging-assisted meniscus detection methods without external reference objects and a submerged continuous water injection method from the pipe orifice are proposed.During the rapid water injection phase in the flask body,the maximum row grayscale detection method is employed for rapid meniscus detection.In the slow water injection phase at the flask neck,either the maximum particle detection method or the edge detection method is used to track the meniscus.During the final constant volume phase,the maximum particle detection method is first used to construct the region of interest for the meniscus.Then,the row grayscale gradient detection method is utilized for precise detection of the meniscus lowest point(concave point).For cases where the concave meniscus is not obvious,the edge detection method is adopted.Experimental verification was conducted using a self-developed machine vision automatic water injection and constant volume device for milliliter-scale volumetric flasks.The results show that the standard deviations for constant volume of 1,2,and 5 ml volumetric flasks are 0.001 9,0.002 0,and 0.008 0 ml,respectively,meeting the requirements of metrological work.关键词
机器视觉/毫升级容量瓶/注水定容/液面检测Key words
machine vision/milliliter-scale volumetric flasks/water injection and constant volume/meniscus detection分类
机械制造引用本文复制引用
彭新华,蔡永洪..基于机器视觉的毫升级容量瓶注水定容方法[J].自动化与信息工程,2025,46(6):17-23,7.基金项目
青海省重点研发与转化计划(科技援青合作专项)项目(2025-QY-254) （科技援青合作专项）
