郭常盈 吴冉

摘 要： 现有的相位测量算法进行光反馈自混合干涉位移实时跟踪测量时，未考虑原始光反馈自混合干涉信号中的噪声干扰，导致测量精度差。因此，对光反馈自混合干涉位移实时跟踪测量算法进行改进，采用自适应滤波系统去除光反馈自混合干涉信号中的噪声，通过去噪后的自混合信号实施变换处理获取解析信号，提取解析信号的包裹相位突变点，根据相位突变点与正负单位脉冲的乘积定位相位突变点的正确方向，得到自混合干涉条纹；去除自混合干涉条纹中的小数条纹获取全部整数条纹点，对条纹点进行赋值、差值和平滑处理，获取外部目标重构位移轨迹，即光反馈自混合干涉位移。实验结果表明，所提算法具有较高的光反馈自混合干涉位移实时跟踪测量精度。

关键词： 光反馈； 自混合干涉； 去噪； 相位突变点； 整数条纹； 位移测量

中图分类号： TN247?34 文献标识码： A 文章编号： 1004?373X（2018）16?0108?04

Abstract： The noise interference in original optical feedback self?mixing interference signals is not considered when the current phase measurement algorithm is used for real?time tracking measurement of optical feedback self?mixing interference displacement， resulting in poor measurement accuracy. Therefore， improvement of the real?time tracking measurement algorithm for optical feedback self?mixing interference displacement is conducted. The adaptive filtering system is adopted to remove the noises in optical feedback self?mixing interference signals. The transform processing of denoised self?mixing signals is conducted to obtain the analytic signals. The mutation points for the wrapped phase of analytic signals are extracted. The correct direction of phase mutation points is located according to the product of the phase mutation point and the positive or negative unit pulse， so as to obtain self?mixing interference stripes. The decimal stripes in self?mixing interference stripes are removed to obtain all the integer stripe points. The stripe points are assigned， interpolated and smoothed to obtain the displacement track of external target reconfiguration， that is， optical feedback self?mixing interference displacement. The experimental results show that the proposed algorithm has high accuracy in real?time tracking measurement of optical feedback self?mixing interference displacement.

Keywords： optical feedback； self?mixing interference； denoising； phase mutation point； integer stripe； displacement measurement

随着激光技术广泛使用，光反馈自混合干涉技术脱颖而出，以高分辨率、非接触测量、精确度高等优点应用于工业中不同测量环节，在物体震动测量、位移测量、速度测量等方面发挥不可替代的作用[1]。实现光反馈自混合干涉位移实时跟踪测量的算法种类繁多，近年来一般采用相位测量法、条纹计数法、相位解卷法等算法。

传统相位测量算法进行光反馈自混合干涉位移实时跟踪测量时，未考虑原始光反馈自混合干涉信号中的噪声干扰，导致测量精度不理想，本文对此加以改进，引入自适应滤波系统对原始光反馈自混合干涉信号实施去噪处理，一方面能够降低后期测量的难度，另一方面提高测量的精确度[2]。……