首页|一种质量法机械式水表串联自动检定装置设计

一种质量法机械式水表串联自动检定装置设计

扫码查看
针对机械式水表读数的现有图像识别方法识别成功率低、效率低的问题,开发了一种动态识别指针(或梅花轮)位置的图像识别方法.对被测水表进行摄像,自动定位、跟踪末位指针或梅花轮转动圈数和角度(坐标),实时图像识别转化为水量读数,结合计量标准器自动测量的实际累积水量,实现水表示值误差自动检定.设计了一套基于图像识别的质量法机械式水表串联自动检定装置,实现 6~10 台同规格DN(15-25)水表同时自动检定;图像识别成功率高于 99.9%,检定装置准确度等级达 0.2 级;与人工检定相比,检定时长从 23.5~43 分钟降低至 11~15 分钟,显著提升了水表检定工作质量和效率.
An Automatic Verification Device for Mechanical Water Meter Series Based on Quality Method
A water meter reading image recognition method based on dynamic recognition of the position of the pointer(or plum blossom wheel)is developed to address the low success rate and efficiency of existing image recognition methods for mechanical water meter readings.By capturing the measured water meter,automatically locate and track the number of turns and angles(coordinates)of the last pointer or plum blossom wheel,convert real-time image recognition into water volume readings.An automatic verification device for mechanical water meter series based on quality method and image recognition has been designed to achieve simultaneous automatic verification of 6~10 DN(15-25)water meters of the same specification.The success rate of image recognition is higher than 99.9%.Compared with manual verification,the verification duration has been reduced from 23.5~43 minutes to 11~15 minutes.

mechanical water meterautomatic verification devicereadingimage recognitionquality methodseries

曾穗松、张圆明、江洁、韦焕基

展开 >

广州能源检测研究院,广东 广州 511447

机械式水表 自动检定装置 读数 图像识别 质量法 串联

广州市市场监督管理局科技项目广州市市场监督管理局科技项目

2022KJ242023KJ21

2024

工业控制计算机
中国计算机学会工业控制计算机专业委员会 江苏省计算技术研究所有限责任公司

工业控制计算机

影响因子:0.258
ISSN:1001-182X
年,卷(期):2024.37(5)
  • 8