用空间光调制器实现全息再现像的实时重构

针对目前实现大视角数字全息图实时动态再现的难点,提出了一种基于液晶空间光调制器用分时再现原理实现大视角数字全息图实时动态显示的新方法。分析了大视角全息图的计算原理和利用空间光调制器进行分时再现的技术原理,并提出了空间多屏拼接的方法。利用经改造的液晶背投影光学引擎系统设计了实验系统,对计算得到的全息图阵列进行了相应的处理以符合原投影系统RGB信号的传输,处理后的全息图阵列由计算机控制以60Hz的频率输出到光学引擎。实验结果表明,可以观察到大视角全息图的动态再现像。     

医学层析图像的计算全息三维可视化

提出了一种医学层析图像的计算全息三维可视化技术。首先研究了层析图像序列的三维信息融合,将层析图像序列的二维信息融合成三维信息,用计算全息图的方法进行三维信息记录。然后结合空间光调制器的结构特性,对全息系统的空间频率、参物光夹角、取样间隔以及全息再现像的再现区域和视角等进行了讨论和分析,并设置了相关参数,使计算全息系统与电子显示系统相匹配。最后用液晶空间光调制器作为全息图显示载体,用计算机控制全息图的实时输出,用雾屏承载三维空间再现像,建立三维图像光电再现与实时显示系统,实现层析图像序列的三维可视化,给出了理论分析与实验结果。     

用于OCT动态聚焦的液晶菲涅尔波带透镜研究

提出了利用液晶菲涅尔波带透镜实现光学相干层析成像(OCT)动态聚焦的方法.根据扭曲相液晶空间光调制器(TN-LCSLM)的光学特性,设计了适用于OCT动态聚焦的菲涅耳波带透镜.利用TN-LCSLM型菲涅耳波带透镜进行了变焦控制实验,焦距实测结果与设计值比较吻合.此外,本文讨论了采用TN-LCSLM型菲涅耳波带透镜实施动态聚焦涉及到的一些问题.     

基于液晶空间光调制器的相息图扫描三维成像

本文提出了一种利用纯相位型液晶空间光调制器(LC-SLM)实现相息图三维显示的方法.该方法以LC-SLM为显示器件,通过相息图的衍射进行三维像的重构.详细研究了相息图的计算及其与LC-SLM参数间的关系,并对再现像像质进行了讨论.为提高显示的空问分辨率,采用分时复用技术,对三维物体进行分组取样,并计算每一分组的相息图,形成分组相息图序列,再现时依次将相息图输入到LC-SLM,利用人眼的视觉残留效应以达到连续扫描三维成像的目的.实验结果表明,该方法实现了三维物体的再现,为三维显示提供了一种有效的方法.     

美国Carnegie—Mellon大学光学信息处理实验室Casasent教授来华讲学记录 第五讲 空间光调制器

所谓空间光调制器是指能够以非相干光或电子信号通过换能器来调制相干光束的器件。这样,在进行光学图样识别和光学信号处理时,就能以相干光的形式输入系统,实时地进行运算。下面分调制器的分类、性能指标和功能;油膜光阀;电子束寻址的DKDP光阀;液晶光阀;以及光DKDP光阀等五个方面逐一介绍。     

实时联合变换相关器目标探测与识别

光学自动图像识别的基本结构是光学相关器,由于光学相关是所有采样点并行处理,运算速度高于计算机千百倍.本文用电寻址液晶EALCD作为空间光调制器SLM,用计算机控制的EALCD并排显示参考信号和实时扫描的目标信号,并用面阵CCD作为平方律探测器记录其联合变换功率谱,通过光学傅立叶变换,输出相关信号,识别目标并确定目标在空间的方位.其特点是,由于用CCD探测功率谱,可以用各种数字处理技术提高系统的信噪比,并能更精确的探测和识别目标.     

基于EALCD的数字全息的研究

采用一种新的数字全息再现方法,将电荷耦合器件CCD与电寻址液晶EALCD相结合实现数字全息的再现.同时将数字图像处理技术与数字全息术相结合,应用MATLAB软件对记录的全息图进行数字图像处理,实验结果表明,该方法简单快速.通过图像增强等方法对全息图进行数字图像处理,在全息再现中,全息干涉图的对比度得到了显著的提高,有利于全息干涉条纹的自动判读.     

Gyrator变换计算全息图及其新型编码方法

基于gyrator变换数值算法和计算全息理论提出了gyrator变换计算全息图,采用罗曼编码方法制作出了在不同变换角度下的gyrator变换计算全息图并实现了相应的数字再现,其再现规律与相关文献报道的gyrator变换全息图的再现规律一致,表明制作gyrator变换计算全息图是可行的。在此基础上提出了一种新的计算全息编码方法即梯形等效面积编码方法,该方法能有效提高在单个抽样单元中进行相位编码时的相位量化级数。应用该编码方法完成了傅里叶变换计算全息图的产生和数字再现的仿真实验,实验表明采用新型编码方法的离轴物体的计算全息图能实现零级衍射成像。采用新型编码方法也实现了gyrator变换计算全息图的产生和数字再现的仿真实验。上述实验结果表明本文提出的新型编码方法是可行的和有效的。     

计算机-光学联合制作宽视角体视全息图

提出了一种用计算机和光学方法相结合制作大视角白光再现全息图方法,该方法充分利用光学方法和计算机制全息图方法的各自特点以达到对任意三维物体的立体图象重构。     

主动式全息显示的全息图快速生成

全息图的生成速度影响了全息三维显示的实用化,大尺寸、大场景全息图的生成尤为困难,为了解决此问题,将人眼跟踪技术和指向光技术用于全息显示,形成主动式全息显示是一种解决动态全息显示的途径。针对主动式全息显示的需要,提出一种基于二步全息算法及空间冗余光波去除的并行全息图生成方案。首先根据二步全息算法计算行列贡献分量,结合空间冗余光波去除原理确定子全息图范围,合理设计基于GPU的CUDA并行计算方案,以实现大尺寸、大场景全息图的快速生成。实验表明这种方法有效可行,二步算法的引入使计算速度在并行计算的基础上再提高10倍左右,空间冗余光波的去除有效克服大场景与空间采样间隔之间的矛盾。     

递归二元搜索型计算全息图的研究

在分析递归二元搜索算法合成计算全息图原理的基础上，计算机编程设计了递归二元搜索型计算全息图，并在实验上研究了计算全息图的恢复象。实验结果表明，递归二元搜索算法可合成任意波前分布的傅里叶变换计算全息图，并具有恢复象质量好，衍射效率高的特点。     

Reconstruction of complex-valued electron density with x-ray in-line holograms

A reconstruction algorithm is proposed to analyze the complex-valued electron-density distribution in an object with x-ray in-line holograms. The real and imaginary parts of the electron density correspond to local variations of x-ray phase shift and absorption, respectively. In the algorithm, the least-squares error of the holograms is iteratively minimized under sign constraints on the real and imaginary parts. The constraints, which are derived from the physical conditions in the interaction between x rays and materials, facilitate robust reconstruction. The reconstruction was applied to holograms of a biological specimen that caused both phase shift and absorption.     

A Contribution to Hologram Imagery

A very simple method for evaluating hologram imagery in a general holographic arrangement with spherical or plane reference and reconstruction waves is proposed. For the hologram imagery approaching aberration-free case the relations providing new possibilities are introduced. The study is not restricted to plane holograms only but one can also study the imaging of objects reconstructed from volume holograms.     

Magnified reconstruction of digitally recorded holograms by Fresnel-Bluestein transform

A method for numerical reconstruction of digitally recorded holograms with variable magnification is presented. The proposed strategy allows for smaller, equal, or larger magnification than that achieved with Fresnel transform by introducing the Bluestein substitution into the Fresnel kernel. The magnification is obtained independent of distance, wavelength, and number of pixels, which enables the method to be applied in color digital holography and metrological applications. The approach is supported by experimental and simulation results in digital holography of objects of comparable dimensions with the recording device and in the reconstruction of holograms from digital in-line holographic microscopy.     

High-fidelity numerical realization of multiple-step Fresnel propagation for the reconstruction of digital holograms

A cascaded Fresnel algorithm for the flexible reconstruction of digital holograms is proposed. Since the fast-Fourier-transform-based numerical realization of the Fresnel integral shows a dependency of its pixel resolution and its computation window size on the propagation distance different from that of the corresponding physical system, the computation window can be smaller than the actual physical diffraction field in the intermediate plane. Consequently, distortions in the final reconstruction may occur. A method is proposed to eliminate such distortion. The validity of this method is shown by both numerical simulations and experimental results.     

Moiré pattern encoded extended fractional Fourier transform security hologram

This paper describes a simple method for making extended fractional Fourier transform (EFRT) based Moiré pattern encoded security holograms. These security holograms contain multifold concealed and encoded anticounterfeit security features that can only be read through a key hologram and periodic patterns in the final reading process. The encoded features in these holograms are concealed and unknown to the counterfeiter. These features are encoded separately for each individual recording in angularly multiplexed extended fractional Fourier transform hologram (EFRTH). The principle of recording and reconstruction of the proposed security hologram along with experimental results are presented.     

Improved-resolution digital holography using the generalized sampling theorem for locally band-limited fields

We describe the recording conditions that, together with the appropriate numerical reconstruction process, permit high-lateral-resolution reconstruction of in-line digital holograms. By high resolution, we mean a resolution that is beyond the Nyquist frequency, which is achieved by common methods. The proposed method is based on a previously reported generalized sampling theory that presents the conditions to precisely reconstruct fields that in certain cases may be sampled with a sampling rate lower than the Nyquist rate. We examine the hologram-recording process in the Wigner space. On the basis of this analysis, we demonstrate a simple high-resolution numerical reconstruction method.     

In-line hologram reconstruction using Hartley transform

In reconstruction of in-line recorded holograms, zero-order and conjugate images appear on the same physical location as the object image. Here we propose a method, new to our knowledge, to separate the object image from the others by using two quadrature phase-shifted holograms. The method uses the Hartley transform and a phase retrieval type of algorithm on the difference hologram.     

Estimation of wavelength difference using scale adjustment in two-wavelength digital holographic interferometry

We propose a method for an estimation of wavelength difference using scale adjustment in two-wavelength digital holographic interferometry. To estimate wavelength difference, two holograms recorded with different wavelengths are reconstructed on the basis of the Fresnel diffraction integral, and pixel sizes in the reconstruction plane, which depend on the wavelength in recording hologram, are analyzed. In the analysis, a zero-padding method and an intensity correlation function are used to adjust pixel sizes in the reconstruction plane and then obtain a wavelength difference given by a difference between the pixel sizes. Theoretical predictions and experimental results are shown to indicate the usefulness of the proposed method in this paper.