机器视觉精密测量中的显微光学聚焦

发布时间：2018-03-19 12:23

本文选题：机器视觉　切入点：显微成像　出处：《光学精密工程》2016年09期 　论文类型：期刊论文

【摘要】：考虑显微光学涉及的聚焦精度对机器视觉精密测量效果的影响,开展了显微视觉环境下对图像聚焦技术综合定量评价的研究。建立了偏移率等系列性能指标,对13组清晰度函数在显微视觉条件下的无偏性、单峰性、分辨力等进行了综合评价,优选出方差函数和Brenner函数分别用于粗聚焦和精聚焦阶段的清晰度计算。建立了分步爬山搜索法,实现了显微自动聚焦。与传统爬山法相比,提出的方法聚焦时间显著缩短,重复精度提高约24%。将建立的自动聚焦与图像测量方法应用于某电液伺服阀衔铁气隙测量中,得到的测量均值与工具显微镜结果相近,而测量标准差可达1.9μm,测量效率也显著提高。最后对伺服阀加电条件下的气隙动力学特性进行了测试,获得了驱动电流-衔铁气隙之间的关系,为在线装配/装调提供了重要依据。
[Abstract]:Considering the effect of focusing precision involved in microscopic optics on the precision measurement effect of machine vision, the comprehensive quantitative evaluation of image focusing technology is studied in the environment of microscopic vision, and a series of performance indexes, such as offset ratio, are established. The unbias, single peak and resolution of 13 groups of definition functions under microscopic vision were evaluated. The variance function and Brenner function are selected to calculate the sharpness of coarse focus and fine focus, respectively. A step by step mountain climbing search method is established to realize microscopic auto-focusing. Compared with the traditional hill climbing method, the focus time of the proposed method is significantly shortened. The automatic focusing and image measurement method has been applied to the measurement of the armature air gap of an electro-hydraulic servo valve. The mean value of the measurement is similar to that of the tool microscope. The measurement standard deviation can reach 1.9 渭 m, and the measurement efficiency is also improved significantly. Finally, the dynamic characteristics of the air gap under the condition of the servo valve charging are tested, and the relationship between the driving current and the air gap of the armature is obtained. It provides an important basis for online assembly / assembly.
【作者单位】：大连理工大学微纳米技术及系统辽宁省重点实验室;
【基金】：国家科技重大专项基金资助项目(No.2013ZX04001091)
【分类号】：TP391.41

【相似文献】