垂直扫描白光干涉轮廓测量的抗振恢复算法  

Anti-vibration Method for Perpendicular Scanning White Light Interferometry Profiling

作　　者：李泽昊 曲通 肖宇 刘晓军[1] Li Zehao;Qu Tong;Xiao Yu;Liu Xiaojun(School of Mechanical Science&Engineering,Huazhong University of Science and Technology,Wuhan 430074,China;不详)

机构地区：[1]华中科技大学机械科学与工程学院,武汉市430074

出　　处：《工具技术》2025年第3期154-160,共7页Tool Engineering

摘　　要：针对垂直扫描白光干涉轮廓测量精度与其垂直扫描干涉图像轴向定位精度的紧密关系,由于物镜驱动器定位精度及外界环境振动等因素的影响,实际轴向位置与其设定值存在一定偏差。本文基于白光干涉测量原理以及激光相移干涉技术,分析像素点干涉信号,提出干涉质量评价模型来排除失真信号的干扰,同时提取各有效干涉信号峰值所在图像帧数作为信号峰值初矩阵,设定一定的宽松系数排除没有干涉条纹的图像或仅包含低对比度的干涉条纹图像。通过循环迭代的方法,从有效干涉图像的第一帧图像开始,将每次计算中信号峰值在该次测量范围内的像素点作为本次计算的有效计算区域,对该区域采用最小二乘法提取局部移相量,最后将局部移相量求解为整体的步长并进行三维表面重建。仿真和实验结果表明,本文所提算法在不同振动环境下步长提取误差小于0.5 nm,在振动环境下的测量精度相比传统算法高,表现出良好的抗振能力。In view of the close relationship between the measurement accuracy of the vertical scanning white light interferometric profile and the axial positioning accuracy of the vertical scanning interferometric image,there is always a certain deviation between the actual axial position and the set value due to the influence of the positioning accuracy of the objective lens driver and the vibration of the external environment.Based on the principle of white light interferometry and laser phase-shift interferometry,through the analysis of pixel interference signals,an interference quality evaluation model is proposed to eliminate the interference of distorted signals,and the number of image frames where the peaks of each effective interference signal are located is extracted as the initial matrix of signal peaks,and a certain relaxation coefficient is set to exclude images without interference fringes or images with only low contrast interference fringes.Through the cyclic iterative method,starting from the first frame of the effective interference image,the pixels in the measurement range of the signal peak in each calculation are taken as the effective calculation area of this calculation,and the local phasor shift is extracted by the least squares method for this region.The local phasor shift is solved as the overall step size and the three-dimensional surface reconstruction is carried out.Simulations and experiments show that the step size extraction error of the proposed algorithm is less than 0.5 nm in different vibration environments,and the measurement accuracy in vibration environments is significantly improved compared with traditional algorithms,showing good vibration resistance.

关 键 词：白光干涉 轮廓测量 抗振 有效计算区域 最小二乘迭代 

分 类 号：TG806[金属学及工艺—公差测量技术] TH164[机械工程—机械制造及自动化]