机器视觉非接触测量系统在线校准及不确定度分析  

On-line calibration and uncertainty analysis for non-contact measurement system based on machine vision

作　　者：宋勇[1] 董铁[1] 吴纪国[1] 

机构地区：[1]中国工程物理研究院机械制造工艺研究所,四川绵阳621900

出　　处：《机械》2014年第10期70-73,共4页Machinery

摘　　要：高精度、大面阵CCD和CMOS传感器件的出现,使得机器视觉非接触测量技术得到越来越广泛的应用。本研究以典型机器视觉非接触测量模型为研究对象,通过对其原理、测量数学模型和计量特性进行了阐述和分析,创新性提出了一种机器视觉非接触测量系统的在线校准方法,分析并确立了测量过程中的不确定度来源,并对校准结果进行了不确定度评定。实验表明,机器视觉非接触测量系统在线校准方法具备两个优点:一是易操作性,优化了测量标准的追溯关系;二是校准精度高,在95%的置信概率下,校准系统不确定度为0.008 mm。With the advent of high precision and large-area CCD and CMOS sensors, non-contact measurement system based on machine vision is used more and more widely. The research takes typical non-contact measurement mode based on machine vision on as the object of the research. Through exposition and analysis for the principle, measurement mathematic model and measurement characteristic of it, the article puts forward a on-line calibration method for non-contact measurement system based on machine vision, which analyses and establishes the source of uncertainty and evaluates the uncertainty of calibration result. The massive experiments indicates that the method is equipped with two advantages. One is handleability, which optimizes the tracing relation. The other is the high precision, the uncertainty of the calibration system can achieve 0.008mm with confidence probability of 95%.

关 键 词：机器视觉 校准 追溯关系 不确定度 

分 类 号：TH741[机械工程—光学工程]