基于模式识别的高效亚微米级不确定度测量方法

High-efficiency sub-microscale uncertainty measurement method using pattern recognition.

作者信息

Zhao Chenyang, Cheung Chi Fai, Xu Peng

机构信息

School of Mechanical Engineering and Automation, Harbin Institute of Technology, Shenzhen 518055, China; State Key Laboratory of Ultra-precision Machining Technology, Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong, China.

State Key Laboratory of Ultra-precision Machining Technology, Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong, China.

出版信息

ISA Trans. 2020 Jun;101:503-514. doi: 10.1016/j.isatra.2020.01.038. Epub 2020 Feb 3.

DOI:10.1016/j.isatra.2020.01.038

PMID:32037053

Abstract

This study presents a fast precision measurement method that uses pattern recognition. First, a specific micro-structured surface was designed and manufactured, providing a unique pattern for recognition and matching. Second, a measurement system was proposed based on the algorithms of circle Hough transform (CHT), neural classifier (NC), template matching (TM) and sub-pixel interpolation (SI). Then, a series of experiments were carried out from three aspects: circle detection, length uncertainty, and measurement speed and range. The results showed the correct circle classification percentage was more than 96% and the CHT search accuracy was within a two-pixel level. The length uncertainty test demonstrated the method was able to achieve 90-nm length uncertainty, and a comparison of measurement speeds showed it helped to speed up measurements by a factor of 1000 compared to the original one.

摘要

本研究提出了一种采用模式识别的快速精密测量方法。首先，设计并制造了一种特定的微结构表面，提供了用于识别和匹配的独特图案。其次，基于圆霍夫变换（CHT）、神经分类器（NC）、模板匹配（TM）和亚像素插值（SI）算法提出了一种测量系统。然后，从圆检测、长度不确定度以及测量速度和范围三个方面进行了一系列实验。结果表明，正确的圆分类百分比超过96%，CHT搜索精度在两像素水平以内。长度不确定度测试表明该方法能够实现90纳米的长度不确定度，测量速度比较显示，与原始方法相比，它有助于将测量速度提高1000倍。