First-order optical systems with unimodular eigenvalues

It is shown that a lossless first-order optical system whose real symplectic ray transformation matrix can be diagonalized and has only unimodular eigenvalues is similar to a separable fractional Fourier transformer in the sense that the ray transformation matrices of the unimodular system and the separable fractional Fourier transformer are related by means of a similarity transformation. Moreover, it is shown that the system that performs this similarity transformation is itself a lossless first-order optical system. Based on the fact that Hermite-Gauss functions are the eigenfunctions of a fractional Fourier transformer, the eigenfunctions of a unimodular first-order optical system can be formulated and belong to the recently introduced class of orthonormal Hermite-Gaussian-type modes. Two decompositions of a unimodular first-order optical system are considered, and one of them is used to derive an easy optical realization in more detail.     

First-order PMD compensation using optical FIR filter

We present a synthesis algorithm to design an optical finite impulse response (FIR) filter for compensating a first-order polarization mode dispersion (PMD) by minimizing the differential group delay (DGD). The desired frequency response was approximated using two widely used methods in designing digital FIR filters: the Fourier series expansion method and the frequency sampling method. A numerical simulation was performed for an eighth-order filter to demonstrate the difference between the two methods. The simulation results produced a sharper cutoff for the Fourier series expansion and higher stopband attenuation for the frequency sampling method. The Fourier series method produced better results in reducing the DGD.     

First-order gradients of skew rays of axis-symmetrical optical systems

Current commercial software for analysis and design of optical systems use finite difference (FD) approximation methodology to estimate the gradient matrix of a ray with respect to system variables. However, FD estimates are intrinsically inaccurate, subject to gross error when the denominator is excessively small relative to the numerator. We avoid these problems and determine these gradients by the application of Snell's law. We give the background and basics for determining the first-order gradients of skew rays of optical systems, whereby the differential vector of any ray can be estimated by the product of the developed gradient matrix and differential changes of system variables. The most important application is for optical design by use of optimization methods where the merit function is defined as the spot size. FD used for such optimization is slow for large systems and subject to inaccuracy. The presented methodology is shown to be accurate and computationally faster than traditional FD. Two illustrative examples are provided to validate the proposed method.     

一种基于方向平滑约束优化的多尺度光流算法

提出了一种采用方向平滑约束的多尺度光流算法.该算法在多尺度的拉普拉斯金字塔框架下,利用差值平方和(SSD)区域匹配方法实现对搜索区域内的流速及其协方差矩阵的估计,然后利用方向平滑约束对估计的流速进行传播和优化,方向平滑约束可以有效地抑制对遮挡和运动不连续处的过度平滑.人工和真实图像序列的仿真实验都获得了满意的结果.     

GCPOF: a novel optical flow estimation algorithm leveraged by sparse ground control points

As a fundamental task in computer vision, optical flow estimation algorithms aim to establish dense pixel correspondences between image frames. This paper presents a novel optical flow estimation framework called GCPOF to handle large displacement and scale variations of scene objects, which appear frequently and pose great challenges in practice. Within the framework of GCPOF, large displacement and scale variations are captured by a new problem formulation leveraged by sparse ground control points. We present detailed theoretical derivation of the solution to the problem based on iterative reweighted least squares. Both qualitative and quantitative evaluations on synthetic and real images demonstrate that GCPOF is able to handle optical flow fields with large displacement and scale variations properly, and it runs significantly faster than relevant optical flow estimation methods.     

Electrical and optical characteristics of Ni doped ZnS clusters

A simple, easy approach to the synthesis of nickel (Ni) doped zinc sulphide (ZnS) clusters is reported. The synthesis of Ni–ZnS clusters involved mixing and drying of zinc acetate, sodium sulphide and acrylic acid in appropriate ratio and adding Ni at proper conditions. These clusters were trapped in polyacrylic acid (PAA) to form a PAA capping to provide stability. The clusters were characterized using high resolution SEM for morphological investigation; XRD for its crystalline nature; photoluminescence (PL) for optical characterization and electrical conductivity measurement. Clusters of Ni–ZnS were formed of the size ~ 10 nm.     

First-order chemical reaction on flow past an impulsively started vertical plate with uniform heat and mass flux

Summary A finite-difference solution of the transient natural convection flow of an incompressible viscous fluid past an impulsively started semi-infinite plate with uniform heat and mass flux is presented here, taking into account the homogeneous chemical reaction of first order. The velocity profiles are compared with the available theoretical solution and are found to be in good agreement. The steady-state velocity, temperature and concentration profiles are shown graphically. It is observed that due to the presence of first order chemical reaction the velocity decreases with increasing values of the chemical reaction parameter. The local as well as average skin-friction, Nusselt number and Sherwood number are shown graphically.List of symbols C concentration - C species concentration in the fluid far away from the plate - C _w species concentration near the plate - C dimensionless concentration - D mass diffusion coefficient - Gc mass Grashof number - Gr thermal Grashof number - g acceleration due to gravity - j mass flux per unit area at the plate - K dimensionless chemical reaction parameter - K _l chemical reaction parameter - k thermal conductivity - Nu_x dimensionless local Nusselt number - dimensionless average Nusselt number - Pr Prandtl number - q heat flux per unit area at the plate - Sc Schmidt number - Sh_x dimensionless local Sherwood number - dimensionless average Sherwood number - T temperature - T temperature of the fluid far away from the plate - T _w temperature of the plate - T dimensionless temperature - t time - t dimensionless time - u ₀ velocity of the plate - U, V dimensionless velocity components inX,Y-directions, respectively - u, v velocity components inx, y-directions, respectively - X dimensionless spatial coordinate along the plate - x spatial coordinate along the plate - Y dimensionless spatial coordinate normal to the plate - y spatial coordinate normal to the plate - thermal diffusivity - volumetric coefficient of thermal expansion - * volumetric coefficient of expansion with concentration - coefficient of viscosity - kinematic viscosity - _x dimensionless local skin-friction - dimensionless average skin-friction     

一种改进的主从节点选举算法用于实现集群负载均衡

随着数据量的不断增长,分布式搜索引擎的出现满足了大数据量的检索性能.现有的主从节点选举策略往往会导致主从节点分配不均而使得查询性能不均衡.现提出一种搜索引擎集群的主从节点选举策略,能保证正常情况下节点的分布均衡.当出现宕机时能重新切换主节点,保证检索功能有效;当宕机的服务器重新上线时,主从节点分布恢复正常,避免了负载不平衡的缺陷,达到合理利用每台服务器的性能并充分发挥集群的性能优势的目的.     

Synthesis of thick optical thin-film filters with a layer-peeling inverse-scattering algorithm

We present an efficient and accurate method for synthesis of optical thin-film structures. The method is based on a differential inverse-scattering algorithm and considers therefore both phase and amplitude reflectance data. We apply the algorithm to the synthesis of filters with arbitrary index layers and two-material filters consisting of only high- and low-index layers. The layered structure is approximated by a stack of discrete reflectors with equal distance between all reflectors. This mirror stack is in turn determined from the desired, complex reflection spectrum by a layer-peeling inverse-scattering algorithm. The complexity of the design algorithm is approximately the same as that of the forward problem of computing the spectrum from a known structure.     

Study of electrical and optical properties of Mn doped ZnS clusters

A simple, easy approach to the synthesis of manganese Mn doped zinc sulphide (ZnS) clusters is reported. The synthesis of Mn–ZnS clusters involved mixing and drying of zinc acetate, sodium sulphide and acrylic acid in appropriate ratio and adding Mn at proper conditions. These clusters were trapped in polyacrylic acid (PAA) to form PAA capping to provide stability. The clusters were characterized using high resolution SEM for morphological investigation; XRD for its crystalline nature; photoluminescence (PL) for optical characterization and electrical conductivity measurement. Clusters of Mn–ZnS were formed of the size ~ 10 nm.     

Comparison of three separated flow models

 An improved stochastic separated flow (ISSF) model developed by the present authors is compared with two other widely used trajectory models, the deterministic separated flow (DSF) model and the stochastic separated flow (SSF) model, in numerical simulations of gas–particle flows behind a backward-facing step. The DSF and ISSF models are found to need only 250 computational particles to obtain a statistically stationary solution of mean and fluctuating velocities of the particles, while the SSF model requires as many as 10,000 computational particles. Apart from comparing the sensitivity of required computational particles for different models, prediction capability of different models on mean velocities, fluctuating velocities and re-circulation region are also compared in this paper. Predicted results of streamwise mean velocity of particle phase agree well with experimental data for all the three models. For the mean fluctuating velocity of the particle phase, predictions using the ISSF model agree well with experiment data, while the DSF and the SSF models have a significant difference. Only the SSF and the ISSF models are capable of predicting re-circulation regions of the particle phase. As a comparison, the ISSF model has a distinct advantage over the other two models both in terms of accuracy and efficiency. Received 20 October 2001 / Accepted 5 February 2002     

Iterative zonal wave-front estimation algorithm for optical testing with general-shaped pupils

An iterative zonal wave-front estimation algorithm for slope or gradient-type data in optical testing acquired with regular or irregular pupil shapes is presented. In the mathematical model proposed, the optical surface, or wave-front shape estimation, which may have any pupil shape or size, shares a predefined wave-front estimation matrix that we establish. Owing to the finite pupil of the instrument, the challenge of wave front shape estimation in optical testing lies in large part in how to properly handle boundary conditions. The solution we propose is an efficient iterative process based on Gerchberg-type iterations. The proposed method is validated with data collected from a 15 x 15-grid Shack-Hartmann sensor built at the Nanjing Astronomical Instruments Research Center in China. Results show that the rms deviation error of the estimated wave front from the original wave front is less than lambda/130-lambda/150 after approximately 12 iterations and less than lambda/100 (both for lambda = 632.8 nm) after as few as four iterations. Also, a theoretical analysis of algorithm complexity and error propagation is presented.     

Multiplicative phase-shifting interferometry using optical flow

Fringe patterns with a multiplicative phase shift among them appear in experimental techniques as photoelasticity and RGB shadow moiré, among others. These patterns cannot be processed using standard phase-shifting demodulation techniques. In this work, we propose to use a multiframe regularized optical flow algorithm to obtain the interesting modulating phase. The proposed technique has been applied to simulated and experimental interferograms obtaining satisfactory results.     

Simulation of optical fiber growth in three dimensions

We have carried out 3-dimensional molecular dynamics simulations of a crystal growth technique known as laser heated pedestal growth. Previously, we simulated this technique in 2 dimensions and our current work is an attempt to extend those simulations to the more realistic 3D case. We demonstrate that such 3D simulations are feasible in a study of the influence of the growth speed and the influence of the laser heating on various aspects of the fiber growth. Although much more realistic than the 2D study, the moderate system sizes used in this study and other crude representations of the experimental process still prevent us from making quantitative comparisons with the latter.     

The optimization on flow scheme of helium liquefier with genetic algorithm

There are several ways to organize the flow scheme of the helium liquefiers, such as arranging the expanders in parallel (reverse Brayton stage) or in series (modified Brayton stages). In this paper, the inlet mass flow and temperatures of expanders in Collins cycle are optimized using genetic algorithm (GA). Results show that maximum liquefaction rate can be obtained when the system is working at the optimal parameters. However, the reliability of the system is not well due to high wheel speed of the first turbine. Study shows that the scheme in which expanders are arranged in series with heat exchangers between them has higher operation reliability but lower plant efficiency when working at the same situation. Considering both liquefaction rate and system stability, another flow scheme is put forward hoping to solve the dilemma. The three configurations are compared from different aspects, they are respectively economic cost, heat exchanger size, system reliability and exergy efficiency. In addition, the effect of heat capacity ratio on heat transfer efficiency is discussed. A conclusion of choosing liquefier configuration is given in the end, which is meaningful for the optimal design of helium liquefier.     

Depth-resolved two-dimensional stokes vectors of backscattered light and mueller matrices of biological tissue measured with optical coherence tomography

Mueller matrices provide a complete characterization of the optical polarization properties of biological tissue. A polarization-sensitive optical coherence tomography (OCT) system was built and used to investigate the optical polarization properties of biological tissues and other turbid media. The apparent degree of polarization (DOP) of the backscattered light was measured with both liquid and solid scattering samples. The DOP maintains the value of unity within the detectable depth for the solid sample, whereas the DOP decreases with the optical depth for the liquid sample. Two-dimensional depth-resolved images of both the Stokes vectors of the backscattered light and the full Mueller matrices of biological tissue were measured with this system. These polarization measurements revealed some tissue structures that are not perceptible with standard OCT.     

一种基于模型的光学层析图像重建方法

光学层析图像重建是个病态问题,测量误差会在重建过程中被放大,对此,提出一种以广义高斯马尔可夫随机场模型为先验信息的光学层析图像重建方法.重建过程是对目标函数的优化过程,目标函数关于光学参数的梯度计算是算法中的难点,因此,提出一种基于梯度树的梯度计算方法.文中分别给出了吸收系数和散射系数的重建结果,并引入三个指标因子衡量重建图像的质量,进而列出不同重建算法下,重建图像的指标值.最后通过对重建结果和指标因子取值的比较,分析基于模型的重建算法的有效性.     

光互连网络中排序算法研究

通过对光互连网络排序算法的研究,提出利用二分法构造二分图依次确定内外节点开关的连接状态,得到可重排无阻塞的 Omega 网络, Banyan 网络和 Crossover 光网络,每种光互连网络都可实现 8×8 信号全排列无阻塞的输出和排序。针对二分法互连函数不一致的问题,继而采用优化的 Looping Algorithm 算法,生成路由标签确定各级节点开关的状态,从而得到互连函数相同结构简单性能优越的光网络。     

Parallel detection algorithm for page-oriented optical memories

We present a parallel algorithm for the reliable detection of two-dimensional binary data in page-oriented memories. The development of the proposed pseudodecision-feedback equalization (PDFE) method is motivated by the classical decision-feedback equalization receiver. The technique takes advantage of the known or the estimated optical system characteristics to mitigate space-variant blur and additive thermal noise. We extend the method to correct for fixed-pattern errors including magnification, rotation, and transverse shift. Advantages of the PDFE algorithm include its parallel design, low computational complexity, and local connectivity. A system-capacity metric is used to compare the performance of the PDFE receiver with other conventional approaches, including the simple threshold, the 1:2 modulation code, and the Wiener filter. Results show the PDFE to outperform all the above techniques over a variety of channels for both incoherent and coherent systems. Implementation issues are discussed, and a MOSIS (Metal-Oxide Semiconductor Implementation Service) 2-mum design is presented.