基于交点法的高精度指针式仪表自动检定系统设计
|
摘要
利用机器视觉的办法对高精度指针式仪表进行自动检定,存在相机与指针所在平面和仪表平面不垂直而产生的视差的问题。针对这一问题,查表法必须要求相机和仪表表盘位置相对固定,但不同仪表的表壳尺寸总有差异,查表法无法消除因仪表尺寸差异所带来的误差而造成的读数不准确;模板匹配结合查表法准确度又有所下降。本文提出一种"交点法"实现相机和仪表指针旋转中心对准。实验表明,采用的方法显著提高读数的准确性。
Automatic calibration of high precision pointer instrument by machine vision method has the problem of parallar caused by the plane of camera and pointer and the plane of instrument not perrendicular to each other. For this problem,the table searching method must be relatively fixed position of the camera and instrument dial,while the shell size of different instruments are always different,the table searching method can not eliminate the error caused by inaccurate readings result of the instrument size differences. The accuracy of template matching combined with table searching decreases again. This paper presents a "cross point method"to make the alignment of the camera and the pointer rotation center. Experiments demonstrate that the accuracy of readings is significantly improved by the proposed method.
Automatic calibration of high precision pointer instrument by machine vision method has the problem of parallar caused by the plane of camera and pointer and the plane of instrument not perrendicular to each other. For this problem,the table searching method must be relatively fixed position of the camera and instrument dial,while the shell size of different instruments are always different,the table searching method can not eliminate the error caused by inaccurate readings result of the instrument size differences. The accuracy of template matching combined with table searching decreases again. This paper presents a "cross point method"to make the alignment of the camera and the pointer rotation center. Experiments demonstrate that the accuracy of readings is significantly improved by the proposed method.
引文
[1]林剑萍,廖一鹏.基于Open CV和LSSVM的数字仪表读数自动识别[J].微型机与应用,2017,36(2):36-40.
[2]邓清男,石晓龙.基于椭圆检测的指针式仪表的读数研究[J].图形与多媒体,2018:166-169.
[3]陈树,王磊.基于机器视觉的指针式仪表检测方法[J].计算机与数字工程,2016,(9):1821-1826.
[4] Garay JRB,Kofuji S T,Tiba T. Overview of a system AMR based in computational Vision and Wireless sensor network[C]. IEEE Latin-American Conference On Communications,2009:48-49.
[5]余世周.一种电力指针式仪表示数自动识别的鲁棒方法[J].计算机技术与发展,2018,(4):622-627.
[6]冯养杰,林小竹.基于改进Hough变换的指针式仪表自动识别方法研究[J].北京印刷学院学报,2015,23(24):62-66.
[7]王高登,钟俊.基于改进的Hough变换的指针识别算法[J].现代电子技术,2016,39(16):23-26.
[8]郭午阳.自动识别技术在物流管理中的应用分析[J].信息与电脑,2017,(4):135-136.
[9]陈昕然.基于机器视觉的电力仪表自动检定系统研究[D].哈尔滨工业大学,2015.
[10]于春书,董静宜.模糊图像中仪表盘修正力Snake模型检测[J].哈尔滨理工大学学报,2018,(2):65-69.
[11]朱玉欣,孙立辉.基于种子填充的指针式仪表自动识别算法[J].电子技术与软件工程,2017,(59):79-81.
[12]邢浩强,杜志岐.变电站指针式仪表检测与识别方法[J].仪器仪表学报,2017:55-58.
[13] Wen He. Automatic Calibration System for Digital-Display Vibrometers Based on Machine Vision[J]. Metrology and Measurement Systems,2014:317-328.
[14]宋伟.基于指针区域特征的仪表读数识别算法[J].仪器仪表学报,2014,(S2):50-58.
[15] Peng Chen. Design and motion tracking of a strip glove based on machine vision[J]. Neurocomputing,2018,(10):20-29.
[16]莫文雄,裴立强等.基于模板匹配和查表法的高精度指针式仪表自动检定系统研制[J].电测与仪表,2017,54(12):100-105.
[2]邓清男,石晓龙.基于椭圆检测的指针式仪表的读数研究[J].图形与多媒体,2018:166-169.
[3]陈树,王磊.基于机器视觉的指针式仪表检测方法[J].计算机与数字工程,2016,(9):1821-1826.
[4] Garay JRB,Kofuji S T,Tiba T. Overview of a system AMR based in computational Vision and Wireless sensor network[C]. IEEE Latin-American Conference On Communications,2009:48-49.
[5]余世周.一种电力指针式仪表示数自动识别的鲁棒方法[J].计算机技术与发展,2018,(4):622-627.
[6]冯养杰,林小竹.基于改进Hough变换的指针式仪表自动识别方法研究[J].北京印刷学院学报,2015,23(24):62-66.
[7]王高登,钟俊.基于改进的Hough变换的指针识别算法[J].现代电子技术,2016,39(16):23-26.
[8]郭午阳.自动识别技术在物流管理中的应用分析[J].信息与电脑,2017,(4):135-136.
[9]陈昕然.基于机器视觉的电力仪表自动检定系统研究[D].哈尔滨工业大学,2015.
[10]于春书,董静宜.模糊图像中仪表盘修正力Snake模型检测[J].哈尔滨理工大学学报,2018,(2):65-69.
[11]朱玉欣,孙立辉.基于种子填充的指针式仪表自动识别算法[J].电子技术与软件工程,2017,(59):79-81.
[12]邢浩强,杜志岐.变电站指针式仪表检测与识别方法[J].仪器仪表学报,2017:55-58.
[13] Wen He. Automatic Calibration System for Digital-Display Vibrometers Based on Machine Vision[J]. Metrology and Measurement Systems,2014:317-328.
[14]宋伟.基于指针区域特征的仪表读数识别算法[J].仪器仪表学报,2014,(S2):50-58.
[15] Peng Chen. Design and motion tracking of a strip glove based on machine vision[J]. Neurocomputing,2018,(10):20-29.
[16]莫文雄,裴立强等.基于模板匹配和查表法的高精度指针式仪表自动检定系统研制[J].电测与仪表,2017,54(12):100-105.