首页 | 本学科首页   官方微博 | 高级检索  
     

指针式仪表示值识别系统研究
引用本文:程琦,赵军,刘宇.指针式仪表示值识别系统研究[J].计量学报,2009,30(Z1).
作者姓名:程琦  赵军  刘宇
作者单位:中国计量学院,浙江,杭州,310018
基金项目:浙江省研究生教育创新示范基地支持 
摘    要:指针式仪表识别系统采用机器视觉技术和数字图像处理算法,实现了对指针式仪表示值的自动识别.使用图像差影法、自适应阈值分割算法和最小二乘法进行圆心拟合;提出区域分割方法及中心投影法,对表盘图像进行分块处理,提取表盘刻度,快速识别出表盘指针所处位置.实验表明,采用该系统的识别值与目测值的最大偏差为0.16μm,最小偏差为0.10 μm,检定时间缩短为原来的1/1000,提高了检定效率、检定精度和检定结果的准确性.

关 键 词:计量学  机器视觉  指针仪表  图像识别  阈值分割  中心投影

Study on the Calibration of Analog Measuring Instrument System
CHENG Qi,ZHAO Jun,LIU Yu.Study on the Calibration of Analog Measuring Instrument System[J].Acta Metrologica Sinica,2009,30(Z1).
Authors:CHENG Qi  ZHAO Jun  LIU Yu
Abstract:The calibration of analog measuring instrument system uses machine vision technology and digital image processing technology to recognize readings of dial gauges automatically. With the improved image subtraction method, adaptive threshold segmen-tation method and the least square method, the circle center will be fitted. A new method named as region-segmentation and central projection for extracting the scale of dial plate was proposed. Then the related errors of the dial gaugs will be estimated according to algorithms established in national calibration regulation. The experiment shows that the system identification value and visual value are 0. 16μm in the maximum deviation, and 0.10μm in the smallest deviation.The system improves the efficiency of test and the accuracy of results.
Keywords:Metrology  Computer vision  Dial gauges  Image recognition  Threshold segmentation  Central projection
本文献已被 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号