指针式仪表的计算机视觉检定系统
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
指针式仪表是一种在工业中广泛应用的仪表,其种类繁多,例如:百分表、千分表、汽车航空仪表等。这些仪表因其操作简便、生产成本低,深受用户的欢迎。但是,指针式仪表也需要在出厂或者经过一段时间的使用之后进行检定,以确定其是否合格。传统上的检定是靠手工操作,通过人眼来识别读数并计算误差的。而这种方法有很多弊端,如:人眼检定对精度的影响较大、检定效率较低等等。因此,需要找出一种具有合适的精度且效率较高的检定方法。
     本论文中,采用了基于计算机视觉技术的检测方法,分析了指针式仪表的表盘特征,设计了表盘指针提取的方案,通过图象处理理论设计系统软件。运用CMOS摄象头以及Windows系统函数实现表盘图象的实时采集,运用步进电机控制仪表指针转动。将图象采集,表盘指针驱动技术与图象处理技术相结合,设计了整套的实时表盘读数、误差比较、检定系统。运用Visual C++编写了软件。最终实现了对指针式仪表的实时检定。
     在精密测量中采用计算机视觉以及图象处理技术是一种比较新的方法,它拥有其他方法所不具有的优点:检定精度、效率都较高。在未来的生产实践中,这种方法必定有更加广阔的应用前景。
As one of the measurement instruments, the gauge with pointer is generally used in the manufacturing process at present. There are many kinks of gauge with pointer including pressure gauge, dial gauge, micrometer gauge, automobile gauge and so on. Because they are convenient for use and have low costs, they are popularly used by people. But they also had to be tested after being made or used for a long time so that to insure they are eligible. Traditionally, the gauges are tested by human eyes. However, some of the subjective factors such as: observation angle, observation distance and fatigue may have effects on the value. This method will lead to not only gross error and small reliability but also low efficiency and body fatigue.
    In this paper, aimed to this laggard condition, the author used the method based on computer vision technology to test the gauges, analyzed the gauges' characters and designed the blue print to pick-up the pointer. The pointer was droved by step motor witch was been controlled though PC parallel port and the images were captured by a CMOS camera. The program was wrote by visual C++ and the testing was realized finally.
    The computer vision and image processing technology being used in the exact measurement is a new method, and it has many good points such as: high precision and efficiency. This method will be widely used in the future.
引文
[1] 赵荣春;数字图象处理导论.西安:西北工业大学出版社,2000.8.
    [2] 章毓晋;图象分割.北京:科学出版社,2001.
    [3] 徐建华,图象处理与分析.北京:科学出版社,1992.
    [4] 阮秋琦:数字图象处理学.北京:电子工业出版社,2001.
    [5] 郑南宁;计算机视觉与模式识别.北京:国防工业出版社,1998..
    [6] 周孝宽;实用微机图象处理.北京:北京航空航天大学出版社,1994.10.
    [7] 王积分,张新荣,计算机图象识别.北京:中国铁道出版社,1988。
    [8] 田捷等,实用图象分析与处理技术.北京:电子工业出版社,1995.
    [9] 竺子民,光电图象处理.武汉:华中理工大学出版社,2000.
    [10] 孙济宇,张秀彬等,Windows下实时图象捕获的两种方法.微型电脑应用,2000.5:40-43
    [11] 孙济宇,张秀彬等,Windows下实时图象捕获的两种方法.微型电脑应用,2000.5:40-43
    [12] 高宗海,林盘等,Windows下视频采集程序的设计.现代电子技术,2002.1:33-35
    [13] 王润生:图象理解.长沙:国防科技大学出版社,1995.10.
    [14] 张海潘,牟永敏,面向对象程序设计实用教程.北京:清华大学出版社,2001.
    [15] 孙宇清,马军,在线检测实时图象处理系统的设计与实现.计算机工程与应用,2000.12:172-173
    [16] 赵红颖等,计算机视觉技术在发动机缺陷检测系统中的应用.光学.精密工程,2000.6:283-286
    [17] 刘文萍,吴立德,图象分割中阈值选取方法比较研究.模式识别
    
    与人工智能,1997.9:271-277
    [18] 王强等,HOUGH变换实时检测算法研究.计算机工程与设计,2001.6:76-80
    [19] 付忠良,一些新的图象阐值选取方法.计算机应用,2000.10:13-15
    [20] 王立强,林斌等,基于USB Video Camera的自动对焦系统.光电工程,2001.10:32-34
    [21] 费耀平等,工业视觉系统及应用.长沙铁道学院学报,1997.3:25-30
    [22] 谢根全,工件表面质量缺陷的计算机自动识别,机械电子,2001.4:51-52
    [23] 朱绍文等,图象理解的应用与发展.计算机工程,1999.10:1-2
    [24] 蒋加伏等,计算机视觉系统研究中存在的问题及解决思路.长沙铁道学院学报,1999.3:13-17
    [25] GALAMBOS, C., MATAS, J., andKITTLER, J. :' Progressive Probabilistic Houghtrans form for line detection' . Processings of IEEE Computer Society Conference on Computer vision and pattern 1999,pp. 554-560 recognition. Los Alamitos, California, June
    [26] PALMER, P. L., KITTLER, J., and PETROU, N. :' Using focus of attention with the Hough transform for accurate line parameter estimation'. Pattern Recgnit., 1994, 27, PP. 1127-1133
    [27] KALVIAINEN,H.,HIRVCNEN,P.,XU,L,and OJA, E. :' Probabilistic and nonprobabilistic Hough transfom-overview and comparisions', Image Vis Comput., 1995, 13, (4), pp. 239-252

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700