非固定相机旋转图像的数字散斑相关计算研究

非固定相机数字散斑相关方法拍摄待测物体表面的目标图像时,所拍摄的目标散斑图像相对于参考散斑图像可能产生位置移动及图像旋转。通过辅助定位的相机标定方法可以计算出被测物体表面与相机的相对位置,但无法获得散斑图像在平面内的旋转角度。数字散斑相关方法对图像旋转较为敏感,随着旋转角度的增加,目标子窗口与参考子窗口的相关性迅速降低。研究图像旋转角度对数字散斑相关方法计算精度的影响,分析相关性随旋转角度降低的原因。对相关搜索方法进行相应的改进,在搜索目标子窗口时增加对子窗口旋转角度搜索,在计算参考图像中子窗口位移量的过程中,同时计算了其旋转角度,从而降低图像旋转的影响,提高数字散斑相关方法的计算精度。     

一种利用图像序列进行摄像机标定的方法

结合消隐点的有关知识,利用图像序列中运动的人体及其阴影的垂直关系,给出了一种实用性比较强的摄像机标定方法.该方法采用线性的方法计算出了可变焦摄像机5参数的内参数矩阵及旋转矩阵,并在一定的条件下计算出平移矩阵,最终将摄像机完全标定.计算内参数及旋转矩阵时不必考虑图像之间对应点之间的关系,从而大大减少了图像检测期间算法的复杂度,提高了标定的速度.此外,模拟和实际图像实验也表明该自标定方法具有较高的鲁棒性和准确性,适用于在线摄像机标定系统.     

旋转不变的非局域均值算法在磁共振图像去噪中的应用

磁共振成像（MRI）实验时常采用多次扫描累加平均提高图像信噪比（SNR）,但当扫描过程中运动引起图像变形时,简单地累加平均就无法奏效.为此,本研究组曾提出一种匹配加权平均方法（MWA）提高图像的信噪比.在此基础上,该文提出一种旋转不变的非局域均值算法（RINLM）,即选取圆形邻域区域并将其划分为一系列以中心像素为圆心的等面积圆环,再计算模式的相似性.RINLM算法可以更好地利用图像中旋转的冗余信息、找到更多的相似结构,提高算法的去噪性能.我们把该方法应用于低信噪比图像序列的平均和去噪中,可以更好地处理旋转的局部运动.与非局域均值算法（NLM）相比,RINLM算法可以进一步提高图像的信噪比;与MWA方法相比,其与RINLM算法的结合可以进一步提高磁共振图像序列信噪比,更好的保持图像边缘信息.     

一种基于先验模板的图像面积特征测量方法*

针对显微视觉中需要反复自动聚焦与自动照明调节以获取高质量图像,降低了微操作效率,提出一种图像特征测量方法.该方法建立在二值图像标准互相关计算模型的基础上.首先,通过设计特殊模板图像,简化该计算模型|通过反推该计算模型,可以在不作预处理基础上,得到目标图像的面积特征参量.利用误差补偿原理,设计两种模板求解两次面积参量,而后求平均值,提高了算法对比度及鲁棒性.强噪音仿真图像实验结果表明：改进方法的测量正确度比常规方法高约20倍|强离焦退化仿真实验表明：改进方法的测量正确度可提高一个数量级.实际图像实验中,改进方法可以在严重离焦以及弱照明条件下有效对图像面积特征进行测量,测量精密度比常规方法高3～5倍.结果表明,该方法在低质量图像面积特征测量中具有实用价值.     

一种基于先验模板的图像面积特征测量方法

针对显微视觉中需要反复自动聚焦与自动照明调节以获取高质量图像,降低了微操作效率,提出一种图像特征测量方法.该方法建立在二值图像标准互相关计算模型的基础上.首先,通过设计特殊模板图像,简化该计算模型;通过反推该计算模型,可以在不作预处理基础上,得到目标图像的面积特征参量.利用误差补偿原理,设计两种模板求解两次面积参量,而后求平均值,提高了算法对比度及鲁棒性.强噪音仿真图像实验结果表明:改进方法的测量正确度比常规方法高约20倍;强离焦退化仿真实验表明:改进方法的测量正确度可提高一个数量级.实际图像实验中,改进方法可以在严重离焦以及弱照明条件下有效对图像面积特征进行测量,测量精密度比常规方法高3～5倍.结果表明,该方法在低质量图像面积特征测量中具有实用价值.     

旋转扫描序列图像的全景图拼接方法

针对火炮身管内表面观测的特殊需求,考虑到传统的图像拼接方法效率低或精度低的缺陷,提出了一种硬件和软件相结合的基于旋转扫描序列内表图像的全景拼接方案和图像拼接算法.该方案用步进电机驱动45.平面反射扫描镜和CCD相机同步旋转来实现对等距内表面序列图像的快速获取,利用获取的序列图像根据步进电机的步进角度完成内表面全景图像的粗级拼接,即在拼接中首先对步进旋转角度θ和图像素距离P进行标定,然后根据标定结果选取待配准的子图像区域,在子图像区域中采用基于相化相关的图像配准方法对图像进行像素级配准,从而完成对序列图像的精级拼接.为了实现图像的平滑过渡避免出现拼接痕迹,最后提出了基于人眼视觉特性的高斯函数权值加权图像融合过渡的方法.试验和仿真结果表明:该方案能够实现内表图像的快速高精度全景拼接,拼接的精度达到像素级,拼接时间小于500 ms,能够满足内腔表面观测的实时性需求,非常适合工程实际应用.     

基于图像锁相的薄膜吸收多通道检测技术

从理论和应用两方面对光学薄膜吸收缺陷的检测方法进行了研究.基于CCD摄像机的网像锁相技术,提出了不同于传统逐点扫描方法的多通道薄膜吸收检测方法.基于逐点扫描法和图像检测法,分别计算了大区域内薄膜单点缺陷所对应的吸收分布,结果证明了图像法应用于较大口径薄膜元件吸收检测的可行性.计算和实验表明:相对于逐点法,图像法测量薄膜速度可以提高上百倍,避免了逐点扫描法中耗时长、效率低的不足,扩展了光热技术在大口径镀膜光学元件中的应用.     

静态傅里叶变换光谱仪补偿板的旋转对干涉图谱的调制研究

在静态傅里叶变换光谱仪中,由于系统装调精度的限制,补偿板相对于分束板会产生一定角度的旋转。补偿板的旋转在横向会导致干涉图像发生错位,在纵向则会导致光程差的改变,致使干涉图像失真。通过对光束在补偿板中传播路径的计算,得到干涉图像的错位量和光程差的变化量与补偿板旋转角之间的函数关系,并根据函数关系计算和分析了补偿板的旋转导致的不同波长干涉图像的混叠和光谱的失真。对于补偿板旋转引入的相位误差,提出离散光谱序列解线性方程组的方法,有效实现了误差光谱的良好校正。     

基于多尺度Radon变换的图像检索

根据Radon变换的统计特性构造了不变量,提出一种新的基于多尺度Radon变换的图像形状检索方法.对检索图像作小波变换,根据小波模极大原理得到边缘图像,对边缘图像构造Radon变换中心矩,在中心矩的基础上根据Radon的统计原理构造出尺度不变矩.由于矩阵的奇异值具有旋转不变性,因此针对不变矩向量矩阵求奇异值,该奇异值特征向量具有平移、尺度和旋转不变性.将该Radon变换的不变量作为形状特征,并进行高斯归一化,按照欧氏距离计算不同图像间的形状相似度.试验结果表明,该方法对高斯噪音具有较强的鲁棒性,与其它方法相比具有较好的检索效果.     

基于特征级融合的多波段舰船目标识别方法

针对复杂情况下海上舰船目标单波段特征识别能力不足的问题,研究可见光、中波红外和长波红外三波段特征图像融合技术,重点解决图像融合方法中存在的算法耗时和融合策略选择的问题,提出了一种新的基于区域协方差矩阵的多波段特征级融合方法,针对可见光图像和红外图像分别设计11维和5维特征向量,协方差矩阵可以将多个特征进行融合,既保证了不同目标之间的区别性,同时又减小计算量。该方法首先利用显著性检测,快速定位图像中的目标区域,然后,针对不同波段图像设计的特征向量定义协方差阵的距离计算公式并进行匹配,通过对图像的一次遍历操作获得积分图像,在协方差计算时达到快速计算的目的,最后利用k-阶最近邻算法对多种舰船目标进行分类识别。利用实拍的3 400余张三波段舰船目标图像作为测试数据。实验主要分为两部分,首先对比单波段和三波段融合识别的识别率,验证所提出的融合方法具有更广的应用范围;然后,在计算效率上对比多种传统的像素级方法,验证采用的特征级融合在计算时间上的优势。实验结果表明,该方法可达到95.1%的识别率,单帧计算耗时约为0.5 s,在实时性和检测率方面都有明显提高。     

Use of spare wavelengths for traffic restoration in a multiwavelength transport network

This article describes three network protection strategies for a single link failure using spare wavelengths in multiwavelength all-optical transparent networks. The use of spare wavelengths is shown to be effective and can be economical when a path restoration technique is used. It is also particularly suitable for high-bandwidth optical crossconnect nodes where the number of input and output ports are generally small. Based on the simulation results, the amount of redundancy required for fill protection is dependent on the protection approach employed. Two algorithms for finding and allocating spare wavelengths are introduced. The algorithms are based on the genetic and heuristic algorithms.     

A state space approach in phase shifting interferometry

The paper proposes a state space approach for the determination of phase distribution in an interferogram in the presence of a non-sinusoidal waveform, random noise, and the miscalibration of the piezoelectric device. In this approach we first estimate the phase step imparted to the piezoelectric transducer (PZT) before determining the phase distribution. A denoising procedure is applied with the knowledge of the state-feedback matrix prior to the extraction of the phase step. This step ensures a good estimation of the phase distribution even at low signal-to-noise ratios. The other salient features of the proposed concept lie in its ability to compensate the detector non-linearity and in the use of non-collimated beams. The comparison of the proposed method with other bench-marking algorithms shows that our approach is efficient and robust. The experimental results show the feasibility of the proposed approach.     

对声波和弹性波传播模拟的Hamilton系统方法

建立了离散化网格上的准粒子体系，引入此体系的Hamilton系统描述，用来模拟声波和弹性波的传播.介绍了准粒子间相互作用的九点模型并给出互作用系数.证明了Hamilton系统方法 和声波、弹性波方程的关系，并给出两个方法中所使用物理量的关系.使用辛算法对给定的 介质模型进行数值模拟. 关键词： Hamilton系统方法 九点互作用模型 声波方程 弹性波方程 辛算法     

水平变化波导中的全局矩阵耦合简正波方法

提出了一种新的水平变化波导中声场的耦合简正波求解方法,该方法能够处理二维点源和线源问题,提供声场的双向解。该方法利用全局矩阵(DGM)一次性求解耦合模式的系数,消除了传播矩阵递推求解中存在的误差累积问题;此外,改善了现有模型中对距离函数的归一化方法,从而避免了泄露模式指数增长导致的数值溢出问题。本文还给出了绝对软海底理想波导中耦合矩阵的闭合表达式,并分析了单个阶梯下简正波耦合现象。此外,本文还计算了理想楔形波导中的声传播问题(ASA标准问题),并与解析解及COUPLE07计算结果进行了比较,结果表明该方法是一种稳定、精确的水平变化波导中的声场计算方法。      

GaAs薄膜的有效量子限制长度及其极化子特性

本文用分数维方法研究Al_xGa_1-xAs衬底上GaAs薄膜中的极化子特性, 提出了确定GaAs薄膜的有效量子限制长度的一个新方法, 解决了原来方法中在衬底势垒处有效量子限制长度发散的困难, 得到了Al_xGa_1-xAs衬底上GaAs薄膜中的极化子的维数和结合能. 关键词： 分数维方法 极化子 GaAs薄膜     

Thermal Field-Flow Fractionation of Initially Dilute Polymer Solutions as a Shear Degradation Model. Scaling Model of Macromolecule Degradation at Concentrations Exceeding the Critical Entanglement Value

Experimental data are presented on thermal field-flow fractionation (TFFF) of anionic polystyrene (PS) samples in the range M = (4–12)·10⁶. They show extensive degradation of macromolecular chains at relatively low rate gradients (G < 30 s^?1). The possibility of the influence of relaxation effects on the shape of fractograms and the elution volumes of the samples was taken into account. Mean concentrations in accumulative zones were evaluated. It was shown that, in all cases when degradation was observed, the accumulative zones are the layers of entangled macromolecules. The use of the scaling approach made it possible to simulate layer extension toward the channel center under the influence of the rate gradient. It was shown that, during stretching, the layer is destroyed into blobs, the size of which is determined by experimental conditions. An expression for the critical gradient leading to layer degradation was derived. Quantitative evaluations of fragment sizes and critical gradients obtained from the model are in good agreement with experimental data. The model developed for specific experimental conditions confirms the proposed general mechanism of the so-called shear degradation of macromolecules. The physical picture of degradation in the TFFF channel was considered.     

Dynamic Analysis of Vibrating Systems with Nonlinearities

The max-min approach is applied to mathematical models of some nonlinear oscillations.The models are regarding to three different forms that are governed by nonlinear ordinary differential equations.In this context,the strongly nonlinear Duffing oscillator with third,fifth,and seventh powers of the amplitude,the pendulum attached to a rotating rigid frame and the cubic Duffing oscillator with discontinuity are taken into consideration.The obtained results via the approach are compared with ones achieved utilizing other techniques.The results indicate that the approach has a good agreement with other well-known methods.He's max-min approach is a promising technique and can be successfully exerted to a lot of practical engineering and physical problems.     

Polygonal Approaches to the Circular Billiard: the Lyapunov Exponent

The Lyapunov exponents for three polygonal approaches to the circular billiard quasiperiodic approach, random approach and equilateral approach are calculated, the chaotic behavior in polygonal billiards is discussed. The role of singularity presented by vertex angles is quantitatively discussed. It is found that for the equilateral approach, the Lyapunov exponents vary with the side number N according to a Poisson law, i.e., λ(N) = aN exp(-βN).     

抛物型方程的演化参数识别方法

给出了一种利用演化计算方法求解微分方程中的参数识别类型反问题的方法。该方法把参数识别问题转化为泛函的优化问题用演化算法来求解,指定待定参数的函数类形式,用遗传算法(Genetic Algorithms)来演化待求参数的最优估计值,并将该方法运用于线性扩散方程和拟线性对流扩散反方程反问题的数值模拟中。     

Single frame blind image deconvolution by non-negative sparse matrix factorization

Novel approach to single frame multichannel blind image deconvolution has been formulated recently as non-negative matrix factorization problem with sparseness constraints imposed on the unknown mixing vector that accounts for the case of non-sparse source image. Unlike most of the blind image deconvolution algorithms, the novel approach assumed no a priori knowledge about the blurring kernel and original image. Our contributions in this paper are: (i) we have formulated generalized non-negative matrix factorization approach to blind image deconvolution with sparseness constraints imposed on either unknown mixing vector or unknown source image; (ii) the criteria are established to distinguish whether unknown source image was sparse or not as well as to estimate appropriate sparseness constraint from degraded image itself, thus making the proposed approach completely unsupervised; (iii) an extensive experimental performance evaluation of the non-negative matrix factorization algorithm is presented on the images degraded by the blur caused by the photon sieve, out-of-focus blur with sparse and non-sparse images and blur caused by atmospheric turbulence. The algorithm is compared with the state-of-the-art single frame blind image deconvolution algorithms such as blind Richardson-Lucy algorithm and single frame multichannel independent component analysis based algorithm and non-blind image restoration algorithms such as multiplicative algebraic restoration technique and Van-Cittert algorithms. It has been experimentally demonstrated that proposed algorithm outperforms mentioned non-blind and blind image deconvolution methods.