Wood elastic characterization from a single sample by resonant ultrasound spectroscopy

The goal of this paper is to propose an experimental method allowing the identification of the complete elastic tensor of anisotropic biological materials such as wood using only one sample. To do so, two complementary methods are used. First, the wood eigen-directions are defined from a sample of spherical shape that is then cut into a cube in a way to perform resonant ultrasound spectroscopy (RUS). The method is successfully applied on a reference beech sample with known orthotropic directions. A comparison of the identified elastic constants with those from the literature and some inferred from ultrasonic transmission measurements is given.     

A Neural Network MCMC Sampler That Maximizes Proposal Entropy

Markov Chain Monte Carlo (MCMC) methods sample from unnormalized probability distributions and offer guarantees of exact sampling. However, in the continuous case, unfavorable geometry of the target distribution can greatly limit the efficiency of MCMC methods. Augmenting samplers with neural networks can potentially improve their efficiency. Previous neural network-based samplers were trained with objectives that either did not explicitly encourage exploration, or contained a term that encouraged exploration but only for well structured distributions. Here we propose to maximize proposal entropy for adapting the proposal to distributions of any shape. To optimize proposal entropy directly, we devised a neural network MCMC sampler that has a flexible and tractable proposal distribution. Specifically, our network architecture utilizes the gradient of the target distribution for generating proposals. Our model achieved significantly higher efficiency than previous neural network MCMC techniques in a variety of sampling tasks, sometimes by more than an order magnitude. Further, the sampler was demonstrated through the training of a convergent energy-based model of natural images. The adaptive sampler achieved unbiased sampling with significantly higher proposal entropy than a Langevin dynamics sample. The trained sampler also achieved better sample quality.     

利用Bayesian-MCMC方法从雷达回波反演海洋波导

应用贝叶斯-蒙特卡罗（Bayesian-MCMC）方法将海洋波导参数的先验信息描述为先验概率密度,结合雷达回波资料（电磁波传播损耗）,得到待反演海洋波导参数的后验概率密度,用马尔可夫链蒙特卡罗（MCMC）-Gibbs采样器采样后验概率密度分布,并用样本最大似然估计值作为对海洋波导参数分布的估计.数值实验结果表明,该方法对先验信息进行了有效利用,反演精度高于遗传算法的反演精度.该方法较为充分利用先验信息,得到解的概率分布,即解的不确定性分析,这在实际应用中有一定的参考价值. 关键词： 波导 电磁波传播损耗 贝叶斯-蒙特卡罗 概率分布     

近红外透射光谱结合规范变量分析用于化学模式分类

应用近红外透射光谱,实施规范变量提取-线性判别分析(CVA-LDA)技术,对其分属不同药物剂量类型进行鉴别。针对药物近红外透射光谱频道变量个数多,彼此间存在严重的复共线性,富有冗余信息,样本类(模式)内的离散度矩阵为奇异,由传统CVA方法提取规范变量的计算不稳健,提出了改进的规范变量分析ICVA方法。它通过嵌入偏最小二乘(PLS)算法,完成规范权矢量的稳健估计,进而用于提取出若干个规范变量。而后,基于规范变量张成的低维空间,构造样本类别的线性判别函数,负责各样本个体类别的判定。实验结果表明,改进的ICVA-LDA方法,克服了LDA对高维小样本数据建模的局限性,模型的判别能力明显优于其他方法。     

Adversarially Training MCMC with Non-Volume-Preserving Flows

Recently, flow models parameterized by neural networks have been used to design efficient Markov chain Monte Carlo (MCMC) transition kernels. However, inefficient utilization of gradient information of the target distribution or the use of volume-preserving flows limits their performance in sampling from multi-modal target distributions. In this paper, we treat the training procedure of the parameterized transition kernels in a different manner and exploit a novel scheme to train MCMC transition kernels. We divide the training process of transition kernels into the exploration stage and training stage, which can make full use of the gradient information of the target distribution and the expressive power of deep neural networks. The transition kernels are constructed with non-volume-preserving flows and trained in an adversarial form. The proposed method achieves significant improvement in effective sample size and mixes quickly to the target distribution. Empirical results validate that the proposed method is able to achieve low autocorrelation of samples and fast convergence rates, and outperforms other state-of-the-art parameterized transition kernels in varieties of challenging analytically described distributions and real world datasets.     

基于贝叶斯分层模型的MCMC方法在闪光图像重建中的应用

针对闪光图像得到的光程数据,采用贝叶斯分层模型建立了后验概率模型,运用Gibbs抽样动态构造马尔可夫链;进而获得了关于线吸收系数的统计结果及其不确定度,并与约束共轭梯度(CCG)方法进行对比分析。数值实验结果表明,马尔可夫链蒙特卡罗(MCMC)方法对理想光程图像的重建结果与真值近似完全一致;在含模糊和噪声时,重建结果与CCG方法相当;当含模糊且噪声干扰较大时,MCMC方法的重建结果要略优于CCG;更重要的是MCMC方法能够给出重建结果的不确定度。     

观测窗中多元非线性时间序列的分析

从观察的角度去了解原有系统的特性，不同程度上会受到观测手段的限制．针对一个观测窗口获得的多元时间序列，利用一种能够更好地体现多元时间序列复杂的时空结构的多元奇异系统分析方法，研究多元序列的相关维数和最大李雅普诺夫指数的变化．利用多层感知器神经网络模型分析其可预测性．研究发现，随着观测窗中多元时间序列的变量数目的增加，反映出的复杂性有所提高，稳定性会产生波动，且相应的可预测性也会产生一些差异． 关键词：     

Measuring flow resistivity of porous materials at low frequencies range via acoustic transmitted waves

An acoustic transmissivity method is proposed for measuring flow resistivity of porous materials having rigid frame. Flow resistivity of porous material is defined as the ratio between the pressure difference across a sample and the velocity of flow of air through that sample per unit cube. The proposed method is based on a temporal model of the direct and inverse scattering problem for the diffusion of transient low-frequency waves in a homogeneous isotropic slab of porous material having a rigid frame. The transmission scattering operator for a slab of porous material is derived from the response of the medium to an incident acoustic pulse. The flow resistivity is determined from the solution of the inverse problem. The minimization between experiment and theory is made in the time domain. Tests are performed using industrial plastic foams. Experimental and numerical results, and prospects are discussed.     

基于单一分光棱镜干涉仪的双通路定量相位显微术

仅仅使用一个单独的分光棱镜(BS),实现了一种用于生物细胞三维成像的双通路定量相位显微术.不同于传统的使用方法,将BS倾斜放置,使中央半反射层与入射光光轴之间存在一个非常小的角度.这样基于BS的分光特性,经过BS后的透射光束和反射光束将会叠加在一起并形成干涉.调节样品位置,利用相机拍摄同时获得了存在π相移的双通路干涉图.这种离轴干涉模式,只需要记录单幅干涉图就可以获得真实的相位信息,方法结构简单,易于操作,适用于微小透明样品的三维形貌测量.     

Geometric Variational Inference

Efficiently accessing the information contained in non-linear and high dimensional probability distributions remains a core challenge in modern statistics. Traditionally, estimators that go beyond point estimates are either categorized as Variational Inference (VI) or Markov-Chain Monte-Carlo (MCMC) techniques. While MCMC methods that utilize the geometric properties of continuous probability distributions to increase their efficiency have been proposed, VI methods rarely use the geometry. This work aims to fill this gap and proposes geometric Variational Inference (geoVI), a method based on Riemannian geometry and the Fisher information metric. It is used to construct a coordinate transformation that relates the Riemannian manifold associated with the metric to Euclidean space. The distribution, expressed in the coordinate system induced by the transformation, takes a particularly simple form that allows for an accurate variational approximation by a normal distribution. Furthermore, the algorithmic structure allows for an efficient implementation of geoVI which is demonstrated on multiple examples, ranging from low-dimensional illustrative ones to non-linear, hierarchical Bayesian inverse problems in thousands of dimensions.     

Deformation of a nanocube with a single incoherent precipitate: role of precipitate size and dislocation looping

ABSTRACT

Precipitate hardening is a key strengthening mechanism in metallic alloys. Classical models for precipitate hardening are based on the average behaviour of an ensemble of precipitates, and fail to capture the complexity of dislocation-precipitate interactions that have recently been observed at individual precipitates in simulations and in-situ electron microscopy. In order to achieve tailored mechanical properties, detailed deformation mechanisms at specific precipitates that account for precipitate size, crystallography, and defect structure must be understood, but has been challenging to achieve experimentally. Here, in-situ scanning electron microscope mechanical testing is used to obtain the compressive stress–strain behaviour at an individual, incoherent Au precipitate within a Cu nanocube, and determine the influence of precipitate and cube size on yield strength and strain hardening. TEM imaging and strain mapping of the initial structure shows misfit dislocations at the Au precipitate, threading dislocations that traverse the Cu shell, and localised and anisotropic strain near the precipitate and threading dislocation. These nanocubes have yield strengths of 800–1000?MPa and strain hardening rate of 1–4?GPa. Yield strength is found to depend on the distance from the precipitate interface to the cube edge, while strain hardening depends on both cube size and precipitate size. An analytical model is developed to quantify the contribution of Orowan looping, Orowan stress, back stress and image stress to plasticity at the Au precipitate. Orowan stress is found to be the largest contributor, followed by back stress and image stress.     

Relaxation times of spin states of all ranks and orders of quadrupolar nuclei estimated from NMR z-spectra: Markov chain Monte Carlo analysis applied to 7Li+ and 23Na+ in stretched hydrogels

The NMR z-spectra of 7Li+ and 23Na+ in stretched hydrogels contain five minima, or critical values, with a sharp "dagger" on the central dip. The mathematical representation of such z-spectra from spin-3/2 nuclei contains nine distinct (the total is 15 but there is redundancy of the ±order-numbers) relaxation rate constants that are unique for each of the spin states, up to rank 3, order 3. We present an approach to multiple-parameter-value estimation that exploits the high level of separability of the effects of each of the relaxation rate constants on the features of the z-spectrum. The Markov chain Monte Carlo (MCMC) method is computationally demanding but it yielded statistically robust estimates (low coefficients of variation) of the parameter values. We describe the implementation of the MCMC analysis (in the present context) and posit that it can obviate the need for using multiple-quantum filtered RF-pulse sequences to estimate all relaxation rate constants/times under experimentally favorable, but readily achievable, circumstances.     

Shell histogram equalization of color images

Histogram equalization (HE) is an effective technique for image enhancement. In this study, we devised a new technique called shell histogram equalization for color images. The technique is a dimensionality reduction method, which transforms 3-D space enhancement to 1-D shell enhancement. First, the 3-D RGB color space is decomposed into L (L = 256) RGB shells, which are similar to a quarter sphere shells or a quarter onion squamae. Then, HE is implemented on shells, and makes the shells coincide with the distribution of the iso-luminance-planes in the RGB cube. After analyzing the computational complexity of the proposed method, comparison experiments are carried out and validated by subjective and objective assessments. The experimental results show that the method provides better enhancement for underexposed and high dynamic range images, and the computational time of the method is much lower.     

铁硅合金中立方织构的形成

铁硅合金(含硅3.25％)的样品,经过不同的冷轧和中间退火程序,在最后高温退火时,可以通过二次再结晶,或者一次再结晶形成集中的立方织构。通过二次再结晶形成立方织构时,并不包括重新形核的过程。(100)[001]取向的二次再结晶“晶核”,在一次再结晶完成后,就已经存在。通过一次再结晶形成的立方织构,由于样品在冷轧后,得到了较强的加工立方织构,退火时通过同位再结晶,就得到了再结晶立方织构。     

基于VHDL技术实现视频采集处理器的控制

针对目前视频图像采集技术中图像采样控制复杂,应用不灵活的问题,基于现有视频采样芯片SAA7111,提出一种采用VHDL技术来模拟实现Ｉ２Ｃ总线接口的方法,控制视频采集处理器实现视频图像采集.实验证明,Ｉ２Ｃ总线控制SAA7111采样图像数据正确、稳定.该方法具有非常好的可移植性.     

Concise presentation of the Coulomb electrostatic potential of a uniformly charged cube

We use a novel method to calculate in closed form the Coulomb electrostatic potential created by a uniformly charged cube at an arbitrary point in space. We apply a suitable transformation of variables that allows us to obtain a simple presentation of the electrostatic potential in one-dimensional integral form. The final concise closed form expression of the Coulomb electrostatic potential of the uniformly charged cube is obtained after completing the calculation of the resulting one-dimensional integrals. Such integrals consist of combinations of products of error functions and power functions that can be solved exactly despite their intimidating appearance. The exact analytic formula for the Coulomb electrostatic potential that we derive reflects the symmetry of the cube and is easy to implement. We illustrate its use by calculating the exact values of the electrostatic potential at some points of symmetry such as the center of cube, center of face of cube, center of edge of cube and corner of cube.     

工业CT切片图像直接生成快速原型层片数据

为了将工业CT与快速原型直接集成,提出了一种由工业CT切片图像直接生成快速原型层片数据的方法。首先对工业CT切片图像进行边缘提取、去噪、细化、连接和轮廓跟踪,获取样件的内外结构轮廓点数据;然后在相邻断层间建立轮廓匹配点对,线性插值层间轮廓数据,以满足快速原型加工对层片厚度的要求,并用三次B样条曲线对所有轮廓进行拟合;最后经过轮廓数据点采样和内外轮廓判断后,生成通用层接口文件输入快速原型系统。试验验证了该方法的有效性与正确性。     

基于Sentinel-2影像与PROSAIL模型参数标定的玉米冠层LAI反演

叶面积指数(LAI)与植被光合作用、蒸腾作用、生物量的形成等有密切联系,是玉米长势监测、灾害胁迫监测、产量预测等重要参数之一,也是辐射传输模型、作物生长模型等机理模型的一个重要参数。Sentinel-2卫星是“全球环境与安全监测”计划的第二颗卫星,具有较高的时空分辨率,且具有红边波段,其可见光和近红外波段的分辨率为10m,是农业遥感应用的理想数据源。PROSAIL辐射传输模型是遥感反演玉米冠层LAI的有效途径,然而在反演中存在输入参数不确定性大、调参困难、病态反演、速度慢等问题。模型的参数标定能够获取观测反射率及不确定性范围内的参数取值,提供丰富准确的参数信息,降低模型反演过程中的偏差。为探索参数标定在玉米冠层LAI反演中的应用,研究以Sentinel-2A卫星影像为数据源,使用马尔可夫链蒙特卡洛方法(MCMC)对PROSAIL模型进行参数标定,通过加入5%的观测光谱不确定性,获取各参数在不确定性范围内的后验取值概率分布,以优化反演过程中的参数设置,提高LAI反演精度。研究结果表明：（1）PROSAIL模型对可见光和近红外波段较为敏感的输入参数有LAI、叶片叶绿素含量及结构系数,将此三个参数作为查找表反演中的可变参数能够有效地进行LAI的反演,反演精度的决定系数达0.7以上。（2）MCMC方法能够对PROSAIL模型进行参数标定,获取研究区内玉米各参数取值分布信息,参数后验分布与实际情况接近,表明利用MCMC方法进行参数标定可行有效。（3）通过参数标定可以有效提高LAI的反演精度,在降低反演偏差和异常值方面尤为明显,参数标定优化后的反演平均偏差由原先的20%降低至8%,同时估算精度由76%提高至90%。研究结果表明：利用MCMC进行PROSAIL模型参数标定,能够提高PROSAIL模型的LAI反演精度,降低反演偏差,为利用PROSAIL辐射传输模型提高作物冠层参数反演精度提供借鉴。     

Coupling control variates for Markov chain Monte Carlo

We show that Markov couplings can be used to improve the accuracy of Markov chain Monte Carlo calculations in some situations where the steady-state probability distribution is not explicitly known. The technique generalizes the notion of control variates from classical Monte Carlo integration. We illustrate it using two models of nonequilibrium transport.     

秧苗状Zn2SnO4纳米线的制备及光致发光性能研究

本文通过化学气相沉积方法成功制备了秧苗状Zn2SnO4纳米线,X射线衍射表明,样品为面心立方结构的Zn2SnO4,同时含有少量的ZnO物相,并且可以通过延长沉积时间的方法获得较为纯净的Zn2SnO4材料.SEM结果显示样品形貌为秧苗状的纳米线.通过样品的光致发光谱的研究发现,Zn2SnO4 的紫外发光强度随着ZnO杂相...