基于快速傅里叶变换的实时相移校正算法

在相移测量剖面术（ＰＭＰ）中,提出了一种基于快速傅里叶变换的相移校正算法,利用最初多采的一幅条纹图的信息和傅里叶变换调制度模板,快速地计算出当前的相移值。采用该算法,能实时准确地控制投影光栅的移动,大大提高了测量系统的工和自动化程度,具有极大的实用价值。     

倾斜相移闭合干涉图的非迭代相位提取方法

针对带倾斜相移误差的闭合干涉图,提出一种非迭代的高精度相位提取方法.该方法用傅里叶变换估计闭合条纹的相位,并用图像分割校正相位的符号,然后利用Zernike多项式拟合确定倾斜相移量,最后用最小二乘拟舍得到高精度相位.数值模拟结果表明:该方法的相位提取误差随着干涉图中条纹根数的增多而减小;当干涉图中条纹根数为4.5时,倾斜相移的估计误差为0.37％.实验结果表明该方法的残余误差均方根值为0.121 7rad.该方法精度高,且无需迭代计算,可应用于相移干涉测量.     

波长调谐干涉测量的相移标定方法

基于菲索干涉的变频相移技术,提出一种分步相移标定方法.生成n组t步相移干涉图,每步相移为π/2;将每步的n组干涉图分别与初始干涉图做相乘和傅里叶变换运算,在此基础上构建每步相移的误差函数;通过求取误差函数极值点来实现对每步相移值的准确标定.结果表明,该标定方法的标定准确度优于传统标定方法,且标定准确度提高了2~3个数量级.该方法不仅避免了传统相移标定中存在的随机误差,也彻底消除了累积误差.     

用双图全息相移干涉法计算与时间有关的位相分布

提出一种双图全息相移干涉法．该方法可以用来计算与时间有关的位相分布，两幅全息干涉条纹图可以同时得到，并且有固定的９０°位相差．虽然两幅条纹图的位相可以随着时间变化，如环境影响，或载荷变化，但他们之间的位相差恒定．该方法利用二维傅氏变换，结合偏振相移法以及分光技术．     

基于复合光栅投影的快速傅里叶变换轮廓术

在实际傅里叶变换轮廓术测量中,获取条纹图的零频分量对傅里叶变换轮廓术的测量精度和测量范围有很大影响,甚至妨碍三维面形的正确重建。π相移技术常被用来消除零频分量对傅里叶变换轮廓术测量的影响,但它需要采集两帧具有π相位差的条纹图。这影响了傅里叶变换轮廓术测量方法的实时性。提出采用复合光栅投影来实现从一帧条纹图中消除零频对傅里叶变换轮廓术测量的影响,该复合光栅是由两个不同频率的载频分别调制与其方向垂直的两帧具有π相位差的条纹并叠加形成的。实验表明,同传统的π相移方法相比,提出的新方法没有明显降低π相移傅里叶变换轮廓术的的测量精度,因此能真正实现实时高速测量。     

相移相位测量的全息再现算法及测量误差分析

用全息原理和方法研究相移相位测量,得到了N步整周期相移再现物光波复振幅同步叠加函数(N步相移函数),同时提出一种新的相移相位测量误差分析和最大误差估计方法。N步相移干涉图是以理想平行光为参考光的无衍射同轴全息图,将其与对应的相移参考光相乘后求和得到N步相移函数;在理想情况下,这是一种复振幅分离、测量和物光波复振幅函数同步叠加方法,存在误差时计算出的相位是最小二乘方法的最佳期望结果。利用N步相移函数得到的N 1步相移函数,说明非理想N步相移函数是理想N步相移函数与误差函数之和,可以把相位型误差转化为与振幅和强度相对误差同等的误差来对待,降低了相位测量中误差估计的难度,给出了N步相移算法最大误差的估计方法和公式。     

两步相移实现投影栅相位测量轮廓术

提出一种新的投影栅相位测量方法--两步相移法。该方法只需两幅相移条纹图，因此计算量小，速度快。给出了实验及计算结果，并同四步相移法进行了比较，证明了该方法具有较高的精度。     

两步广义相移干涉术的光学实验验证

提出了一种基于两步广义相移干涉术的波前再现技术以及基于衍射场统计特性的未知相移值提取算法,并且进行了光学实验验证.该方法中可仅利用两幅干涉图及一次不需精确预设的未知相移,参照物光光强和参考光光强,即可通过计算提取真实的相移量,进而恢复出物波波前复振幅,而无需借助傅里叶谱分析或数字滤波等操作.在光学实验中,对借助于空间光调制器（SLM）的间接输入和三维漫反射物体直接输入两种情况进行了详细地研究,得到的光学实验结果验证了所提方案的有效性和可行性. 关键词： 信息光学 相移干涉术 波前再现 数字图像处理     

基于迭代优化的两帧随机相移干涉术

提出了一种由粗及精的无需预滤波的两帧先进迭代随机相移技术。基于最小化残差背景，首先利用采集条纹图的相关运算，在近似正交条件下估计相移和相位；然后利用这些求解过程中的中间参数在空域由条纹图数据估计相移和背景；最后，迭代上述过程直至收敛到预定义的精度，从而实现相位的精确求解。针对条纹图空域背景和振幅变化问题，利用条纹图子图降低背景空域变化对背景和相移估计的影响，并减小运算量。与现有的方法相比，本文方法由于提供了正确的初值，具有收敛快、计算快的特点。本文方法简单可靠，是一种实用的两帧相移干涉方法。     

基于条纹等值线相关的散斑条纹图处理新方法

在数字散斑干涉术中,通常由物体变形前后得到的两幅原始散斑场用相减方法来得到条纹图,并由相移后的多幅(三幅以上)条纹图来求取相位场。介绍了一种基于条纹等值线相关的条纹图生成新方法,并在此基础上提出了基于单幅干涉条纹图的相位场提取新方法和基于条纹等值线相关的单步相移新方法。实验结果表明,该方法具有良好的结果和发展前景。     

基于傅里叶变换的高精度条纹细分方法

针对传统傅里叶变换法在提取条纹图相位中存在的能量泄漏问题,提出了条纹图整周期裁剪的方法,可有效抑制能量泄漏对检相精度的影响,提高傅里叶变换法相位计算的精度。在此基础上,提出了一种基于傅里叶变换时移特性的叠栅条纹细分新方法。与传统傅里叶变换法相比,该方法求取相邻两帧条纹图间的相移,只需经过两次傅里叶变换,不需要截取条纹图的基频再逆变换回空域,因此计算量至少减少了一倍,计算速度大大提高。数值计算结果表明,对两束单色平面波形成的条纹,理想条件下细分精度高达10-12量级;对高斯包络调制的条纹,细分精度至少可达10-3量级。     

Generalized phase-shifting algorithm for dynamic phase measurement in electron-wave interferometry

A generalized phase-shifting algorithm is implemented in the electron-wave interferometry with a transmission electron microscope, based on simultaneous measurement of the initial phases of the interferograms. The initial phase of each interferogram that has inherent linear carrier fringes is calculated using the Fourier co-efiicients of the interferogram itself corresponding to the carrier-frequency. Taking advantage of this phase-shifting method in both spatial resolution and precision, the phase distribution of a biological weak phase object is successfully measured. This promises to be one of the most spectacular application fields for this new technology.     

Rapid phase extraction method based on the difference image between interferograms with unknown phase-shifting amount

In phase-shifting interferometry, based on the difference image between interferograms with unknown phase-shifting amount, a rapid phase extraction method is proposed. In this method, first, by means of the simple subtraction operation between interferograms captured in one phase-shifting period, a sequence of difference images are generated easily; second, to decrease the random phase-shifting errors induced during the phase-shifting procedure, the summation operation is performed for these difference images; third, based on one time phase recovery operation and one time phase-unwrapping operation, the measured phase with high precision can be obtained rapidly. Moreover, in the proposed method, by means of a simple phase summation operation for all pixel-phases in one interferogram, the corresponding phase-shifting amount of interferogram can be determined conveniently.     

用伽柏小波作干涉波前重建

在分析傅里叶相位展开法中存在的一阶频谱定位误差的基础上,尤其针对实际工作中非严格均匀参考条纹干涉图的相位展开问题,提出基于伽柏小波的局域匹配滤波方法,构造多通道伽柏滤波器,得到在Ｍａｌｌａｔ极大模意义下的最佳匹配相位。     

基于二维光栅分光的同步移相干涉测量技术

为了干涉测量的抗振目的,提出了一种新的同步移相干涉测量方案并搭建了实验装置。整个测量系统在迈克耳孙偏振移相干涉仪的基础上,利用一个正交的二维光栅产生对称分光,选取对于理想光栅衍射效率一致的(±1,±1)级衍射光作为测量分光路,使之分别通过偏振方向依次相差45°的一个偏振片组,从而分别形成0°、90°、180°和270°相移的四幅移相干涉图,按照传统的四步移相算法,对被测波面进行了复原。分析了光强畸变和移相误差对系统的测量误差的影响。利用该系统测量一球面系统,结果与在ZYGO干涉仪上相比较,球面系统的均方根误差相差0.012λ,峰谷值相差0.051λ。     

基于二维傅里叶变换的单帧干涉图相位提取方法

针对大口径光学元件干涉测试过程中,测试装置和干涉腔长较大,气流扰动和环境振动对移相测试过程产生影响等问题,采用一种基于二维傅里叶变换的单帧干涉图处理方法,只需要对一幅空间载频干涉条纹图进行处理即可获得待测相位,具有抗振测试的优点。对该方法的基本原理和算法过程进行分析,并对近红外大口径移相平面干涉仪中600 mm口径的光学平晶进行了面形测试。实验结果表明:采用该方法所得波面峰谷值(PV)为0.112,波面均方根值(RMS)为0.014,与移相算法所得波面数据相比,波面峰谷值偏差不到(1/500);波面均方根值(RMS) 偏差几乎为零。     

Measurement of dynamic crack tip displacement field by speckle photography and interferometry

Speckle interferometry has been used to measure the transient displacement field around a crack due to stress wave loading. Double exposure interferograms are recorded with a pulsed ruby laser, and correlation fringes formed by spatial filtering of the developed film. A new technique is presented for reducing the noise level in spatially filtered speckle interferograms. Independent fringe patterns, obtained from one interferogram but with the filtering aperture sampling different regions of the Fourier transform plane, are digitised by a CCD camera and combined numerically. It is demonstrated that accurate analysis of fringe patterns by the Fourier transform method can be carried out, even if no carrier fringes are present. The measurements are compared with those obtained by speckle photography. Least-squares fitting of the theoretical displacement field is used to calculate the stress intensity factor at the crack tip as a function of time.     

Hybrid two-dimensional continuous wavelet transform for analysis of phase-shifted interferograms

In interferogram or fringe pattern analysis, a technique capable of automatically analyzing interferograms that contain complex fringes and high-level noise as well as defects is highly demanded. In this paper, a hybrid technique combining the concepts of the two-dimensional continuous wavelet transform technique and the phase-shifting technique is proposed. The novel technique takes the advantages of the two existing techniques, and has the ability to accurately and automatically analyze multiple phase-shifted complex interferograms involving noise and defects. The validity of the technique is demonstrated by both computer simulation and real experiment.     

Three-step self-calibrating generalized phase-shifting interferometry

An accurate and fast three-step self-calibrating generalized phase-shifting interferomertry (SGPSI) is proposed. In this approach, two new phase-shifting signals are constructed by the difference interferograms normalization and noise suppressing, then the unknown phase shift between the two difference phase-shifting signals is estimated quickly through searching the minimum coefficient of variation of the modulation amplitude, a limited number of pixels are selected to participate in the search process to further save time, and finally the phase is reconstructed through the searched phase shift. Through the reconstruction of phase map by the simulation and experiment, and the comparison with several mature algorithms, the good performance of the proposed algorithm is proved, and it eliminates the limitation of requiring more than three phase-shifting interferograms for high-precision SGPSI. We expect this method to be widely used in the future.