三元复合表面膜反射光谱的计算

将二元的金属－介质非均匀复合物的等效介质理论推广到三元的情况，并由此计算出两种铝阳极氧化着色谱试样的反射谱。与实验结果对比可看出，计算曲线与实验曲线大致相符，表明该计算方法在一定范围和条件下的适用性，从而为多元复合表面材料光特性的理论研究提供了一种可行的方法。     

基于XGBoost的铝合金LIBS光谱分类识别方法

铝合金作为重要的金属材料,广泛应用于各领域,但大量的铝合金废料却难以进行分类回收。二次资源的回收利用是我国工业绿色、可持续发展的助推器,如何快速、简便地对铝合金废料进行识别分类则成为了铝合金废料回收利用的先决条件。激光诱导击穿光谱（LIBS）是近年来发展快速的一种分析技术,具有快速、全元素分析、实时、原位、远距离检测等优点,已广泛应用于塑料、土壤、肉类、钢铁等识别研究,大多采用最小二乘判别分析法、簇类独立软模式、人工神经网络、支持向量机、随机森林等算法来建立模型。基于迭代型树的XGBoost算法具有正则化、并行处理运算、内置交叉验证和高度的算法灵活性等优势,其模型结构相对简单、运算量较小,且准确率较高,成为近年来机器学习中极受欢迎的算法,因而被广泛应用。基于六种铝合金样品的600组光谱数据,根据NIST原子发射光谱数据库进行光谱特征提取,确定光谱特征谱线的分类依据。利用XGBoost算法进行自动分类及排序,将处理后的光谱数据随机划分为训练集和测试集,通过训练集构建算法模型,提取其分类特征;利用测试集检验模型的稳定性和可用性,防止出现过拟合。XGBoost在固定参数下得到的模型具有一定的自适应性,较少受数据集的影响,总体准确率可达96.67%。其分类特征与已知的元素含量信息相吻合,证明了基于光谱的特征谱线数据,可为分类识别提供参考;同时还可根据XGBoost生成的特征评分来对光谱谱线特征的重要性进行排序。实验结果表明,LIBS可用于不同种类铝合金的快速识别,为废弃金属的分类回收提供了一种新的技术。     

结合区域提取和改进卷积神经网络的水下小目标检测

针对水下小目标信息量有限而难以提取有效特征导致的检测性能不佳问题,提出了一种结合区域提取和融合Hu矩特征的改进卷积神经网络水下小目标检测方法。该方法包含区域提取和分类两个步骤。首先以马尔可夫随机场分割算法为基础进行区域提取,对潜在目标定位的同时降低伪目标对后续分类的干扰;然后提取潜在目标区域的Hu矩特征并融入卷积神经网络,形成一种形状特征表征能力更强的改进卷积神经网络用于分类。声呐实测数据处理结果表明,该方法可以有效提升对水下小目标的发现概率和正确报警率,与其他目标检测方法相比,该方法具有更好的检测性能和泛化性。     

Artificial polarization anomalies from holographic gratings

An aluminum reflection grating made holographically was found to display polarization anomalies when illuminated with p-polarized light. Additional anomalies were detected for both p- and s-polarizations after the grating was coated with a layer of dielectric (410 nm of photoresist). These new anomalies have been interpreted in terms of the optical guided modes of a thin film dielectric waveguide bounded on one side by metal and on the other by vacuum.     

Long-range surface plasmon polaritons propagating on a dielectric waveguide support

In the traditional long-range surface plasmon geometry, an ultrathin metal film is sandwiched between two layers having identical dielectric constants. Here we demonstrate the long-range surface plasmon polariton (LRSPP) properties for a new structure where a thin layer with a dielectric constant exceeding that of the surroundings is inserted within the sandwich, provided the layer thickness d satisfies the condition k(⊥)d=mπ where k(⊥) is the component of the guide wavevector perpendicular to the layer and m is an integer. The resulting plasmon modes have smaller losses and nearly the same phase velocity as the original LRSPP. This provides a strategy to support silver films having thicknesses of 10's of nanometers to create plasmonic devices for sensor applications.     

改进粒子群优化算法的高光谱波段选择

高光谱图像具有数百个连续、狭窄的光谱带,光谱范围跨越可见光到红外光,可提供地物的精细光谱属性,对于地物材质和属性的识别分类具有重要应用价值。针对感兴趣目标选择有限的光谱波段进行传输和处理,对于提升高光谱数据处理时效性、以及设计面向特定应用的实用化光谱仪都具有重要意义。而如何结合目标特征选择最优波段成为在提升处理效率的同时保证目标识别或分类精度的必然要求。因此如何从数以百计维度的高光谱图像中选择出具有较好分类识别能力的波段子集是急需解决的问题。提出基于改进粒子群优化算法的高光谱波段选择方法,该方法区别于传统的粒子群优化算法,引入 “概率突跳特性”,并设定新解的淘汰机制,将“停滞”的新解进行淘汰,提高了算法的全局寻优性能。然后基于目标光谱特征采用了最优波段选择的优化目标函数,通过改进的粒子群优化算法求解目标函数,并将选定的波段子集反馈到支持向量机（SVM）中执行分类应用。采用两个标准的高光谱数据集（Indian Pines, Salinas）对选择出的波段子集进行分类测试,结果表明该方法相较于现有方法具有较高的分类精度,在几种方法中,传统的粒子群算法筛选出的波段效果最差;该算法筛选出的波段的分类精度最好,两个数据集的分类精度分别可以达到98.141 4%和99.084 8%。     

一种新的航天器外露介质充电模型

为综合考虑高能电子辐射与周围等离子体对航天器外露介质充电的影响,在航天器内带电模型的基础上,通过添加边界充电电流来考虑等离子体与航天器介质表面的相互作用,并统一参考电位为等离子体零电位,建立了航天器外露介质充电模型,给出了新模型的一维稳态解法,并与表面充电模型和深层充电模型进行了对比分析.结果表明:新建模型能够综合考虑表面入射电流、深层沉积电流和传导电流对充电的耦合作用过程,实现外露介质表面和深层耦合充电计算,有利于全面评估航天器外露介质的充电问题.     

Study of nanodispersed aluminum and iron alcosols by photoacoustic spectroscopy

Nanodispersed aluminum and iron alcosols were prepared by ultrasonic dispersion of nanodispersed aluminum and iron powders in absolute ethanol. The photoacoustic signal (PAS) produced in modulated CO₂ laser irradiation (1.026 and 1.096 kHz) of alcosols depends on the nature and method of nanoparticle fabrication and does not depend on their concentration in ethanol (within 1-5 g/l). Chemical interaction between metal nanoparticles and ethanol activated by laser irradiation or/and ultrasound is considered as the cause of the PAS.     

Self-consistent LDA calculation in ion neutralization at metal surfaces

The neutralization of He⁺ scattered off aluminum is calculated via a self-consistent LDA where the metal surface is modeled by an LDA jellium surface, and its structure factor is consistently calculated. This approach includes Auger and plasmon-assisted neutralization channels of He⁺ to the He ground state in front of aluminum. We analyze these neutralization channels, which leads us to a revision of the usual calculations of ion neutralization on surfaces depending on the transferred energy lying below, near, or above the metal plasma frequency. The results of this calculation are compared with those of other methods, namely usual unscreened calculations, calculations which extrapolate bulk results, calculations performed for a step potential surface, and surface calculations in the long-distance limit.     

Genetic optical design for a compressive sensing task

We present a sophisticated optical design method for reducing the number of photodetectors for a specific sensing task. The chosen design parameter is the point spread function, and the selected task is object recognition. The point spread function is optimized iteratively with a genetic algorithm for object recognition based on a neural network. In the experimental demonstration, binary classification of face and non-face datasets was performed with a single measurement using two photodetectors. A spatial light modulator operating in the amplitude modulation mode was provided in the imaging optics and was used to modulate the point spread function. In each generation of the genetic algorithm, the classification accuracy with a pattern displayed on the spatial light modulator was fed-back to the next generation to find better patterns. The proposed method increased the accuracy by about 30 % compared with a conventional imaging system in which the point spread function was the delta function. This approach is practically useful for compressing the cost, size, and observation time of optical sensors in specific applications, and robust for imperfections in optical elements.     

Analysis of Frequency Selective Surfaces with Metallic and Dielectric Losses at Millimeter Wave Range

In this paper, a theoretical analysis of scattering from a doubly frequency selective surfaces (FSS) with dielectric and metallic losses in the millimeter wave range is presented. The theoretical analysis involves the solution of the electric field integral equation relating the induced current in the FSS in the presence of anisotropic dielectric layers. The method of moment is employed to obtain numerical results. The frequency selective surface structure considered is composed by conducting patch elements sandwiched between two anisotropic dielectric layers. Three different anisotropic materials are considered: Epsilam-10, Alumina, and Sapphire. The sapphire presents low losses when compared with the other two dielectric materials. The most common metals, as copper and aluminum, are considered in this analysis. None significant metallic losses were observed for any considered metal. Numerical results are presented for the scattering characteristics and for dielectric and metallic losses parameters.     

Designing high k dielectric films with LiPON—Al2O3 hybrid structure by atomic layer deposition

A large amount of ultra-low-power consumption electronic devices are urgently needed in the new era of the internet of things, which demand relatively low frequency response. Here, atomic layer deposition has been utilized to fabricate the ion polarization dielectric of the LiPON-Al₂O₃ hybrid structure. The LiPON thin film is periodically stacked in the Al₂O₃ matrix. This hybrid structure presents a frequency-dependent dielectric constant, of which k is significantly higher than the aluminum oxide matrix from 1 kHz to 200 kHz in frequency. The increased dielectric constant is attributed to the lithium ions shifting locally upon the applied electrical field, which shows an additional polarization to the Al₂O₃ matrix. This work provides a new strategy with promising potential to engineers for the dielectric constant of the gate oxide and sheds light on the application of electrolyte/dielectric hybrid structure in a variety of devices from capacitors to transistors.     

Theoretical study of the normal-incidence reflective spectra for determining the thickness and refractive index of dielectric film

We theoretically study and propose the design concept of a new non-contact measurement method of the thickness and refractive index of dielectric film. The needed measurement set-up includes a probe, an optical detection head and a spectrometer. The probe is composed of two patches of dielectric substrates, and the optical detection head is used to detect the reflective spectra. After depositing a testing dielectric film on the two patches, the coating film thickness and refractive index can be determined by analyzing the normal-incidence reflective spectra of the two dielectric substrates. This method has two requirements. First, the coating dielectric films on the two substrates must be the same. Second, the two dielectric substrates respectively have higher and lower refractive indices than the testing dielectric film.     

Dielectric properties of P2O5-Na2O-Li2O glasses containing WO3, CoO or Fe2O3

The aim of this work was to study some polarization mechanisms in phosphate glasses containing different transition metal oxides, performed at radio frequency (100 kHz) and in the microwave region (9 GHz). Cobalt, iron and tungsten oxides were chosen for the present investigation. The results show that the dielectric constant increases linearly with increasing Co concentration, whereas for iron and tungsten ions, the dielectric constant values were found to decrease up to a given concentration and further, increase for higher concentration of W and Fe. This anomaly was observed both at 100 kHz and 9.00 GHz and was attributed to the valence state of the transition metal ions in the glass structure. The dielectric constant is lower in the microwave region, due to structural relaxation mechanisms.     

Role of precursor alloy phases and intermediate oxides in the preparation of Raney and Urushibara iron

⁵⁷Fe Mössbauer spectroscopy and scanning electron microscopy measurements of precursor phases formed during catalyst preparation and of the catalysts, themselves, demonstrate that the preparation of Raney iron from iron aluminum alloys involves the formation of Fe(OH)₂ and Fe₃O₄ as intermediate phases. The metallic Fe is formed from subsequent reduction of Fe₃O₄ by hydrogen generated by the oxidation of aluminum metal by hydroxide ions. Precursors to Urushibara iron U?Fe (III) are found to consist of Fe?Zn alloys when Zn is used as a reductant and of epitaxial deposits of Fe on aluminum when Al is the reductant. The material resulting from the reduction of the iron salt by aluminum is not a hydrogenation catalyst; the absence of catalytic activity is related to the absence of any alloying of the iron and aluminum. A consideration of the preparation of Raney iron, Urushibara iron, ammonia synthesis and Fischer-Tropsch catalysts leads to the conclusions that catalytic activity is highly correlated to the existence of intermediate mixed-crystals phases and the presence of intimate mixtures of at least two phases in the final catalyst.     

Development a Spectrophotometric of Fe(Ⅲ),Al(Ⅲ)and Cu(Ⅱ)Using Eriochrome Cyanine R Ligand and Assessment of the Obtained Data by Partial Least-Squares and Artificial Neural Network Method—Application to Natural Waters

Simultaneous determination of heavy metal cations and accurate quantitative prediction of them are of great interest in analytical chemistry. This work has focused on a comprehensive comparison of partial least squares (PLS-1) and artificial neural networks (ANN) as two types of chemometric methods. For this purpose, aluminum, iron and copper were studied as three analytes whose UV-Vis absorption spectra highly overlap each other. Accordance with determined parameters (ligand concentration, pH, waiting times, the relationship between absorbance and concentration of metal ion effect and foreign ions) are provided and the optimum conditions. After establishing the optimum conditions for Fe3+, Al3+ and Cu2+ containing mixtures spectrophotometric determinations and the data calibration method of least squares (PLS-1) regression, and artificial neural network (ANN) methods were used. Chemometric methods are applied in a fast, simple, and the results are applicable.     

Brain MRI tissue classification based on local Markov random fields

A new method for tissue classification of brain magnetic resonance images (MRI) of the brain is proposed. The method is based on local image models where each models the image content in a subset of the image domain. With this local modeling approach, the assumption that tissue types have the same characteristics over the brain needs not to be evoked. This is important because tissue type characteristics, such as T1 and T2 relaxation times and proton density, vary across the individual brain and the proposed method offers improved protection against intensity non-uniformity artifacts that can hamper automatic tissue classification methods in brain MRI. A framework in which local models for tissue intensities and Markov Random Field (MRF) priors are combined into a global probabilistic image model is introduced. This global model will be an inhomogeneous MRF and it can be solved by standard algorithms such as iterative conditional modes. The division of the whole image domain into local brain regions possibly having different intensity statistics is realized via sub-volume probabilistic atlases. Finally, the parameters for the local intensity models are obtained without supervision by maximizing the weighted likelihood of a certain finite mixture model. For the maximization task, a novel genetic algorithm almost free of initialization dependency is applied. The algorithm is tested on both simulated and real brain MR images. The experiments confirm that the new method offers a useful improvement of the tissue classification accuracy when the basic tissue characteristics vary across the brain and the noise level of the images is reasonable. The method also offers better protection against intensity non-uniformity artifact than the corresponding method based on a global (whole image) modeling scheme.     

Effect of the gaseous medium in the process of rolling on the microhardness of aluminum and iron

The load dependence of the microhardness of polycrystalline aluminum and iron specimens produced by rolling in a nitrogen, helium, or air medium has been investigated. It has been found that nitrogen and helium have different effects on the microhardness of these metals in the low-load range. This difference is associated with the specific features in the intensity of dynamic penetration of nitrogen and helium into the surface layer of aluminum and iron, which depends on the initial defect crystal structure of the metals, as well as on the type of bonding of helium atoms and nitrogen molecules with metal atoms. It has been shown that the effect of the gaseous medium of the rolling on the microhardness manifests itself only in a very thin surface layer of metal specimens, where the microhardness exhibits a size effect, and an increase in the microhardness indentation depth remains unchanged with an increase in the load and does not depend on the gaseous medium of the prerolling of the specimens.     

On the propagation characteristics of metal clad waveguides with a quantum well

The electromagnetic modes of planar metal clad dielectric waveguides containing an n-doped quantum well (QW) are studied theoretically. Special attention is paid on the coupling between metal surface plasmons and intersubband plasmons and the manifestation of this coupling in the propagation characteristics of metal/QW/dielectric and multimode metal/QW/dielectric/metal waveguide structures. The results obtained indicate that the modification of the propagation characteristic induced by the above-mentioned coupling is substantial only in the case of metal/QW/dielectric waveguide structures.     

基于分块图像小波相对能量和熵的视频镜头边界检测算法

提出了对检测窗内图像帧进行分块并进行小波变换,计算尺度j上分块图像的低频、水平、垂直和对角信号的小波相对能量和小波熵的方法,给出了分块图像小波熵的计算公式.采用尺度j上分块图像低频信号小波熵作为特征向量检测视频镜头是否发生转换,即在检测窗内,帧间低频信号小波熵值只发生一次大突变（与设定的阈值相比）就是剪切转换,若发生多次比较大的变化就是渐变转换.采用检测窗内分块图像在尺度j上的垂直、水平和对角高频信号的小波熵作为另一特征量可以区分渐变镜头的类型：淡进/出(fade in/out)或扫换(wipe).采用大量的高清和标清码流进行实验并与基于概率的方法进行比较.结果证明：本文的视频镜头检测方法具有强的鲁棒性、能检测出镜头转换的类型：剪切、淡进/出和扫换,同时有较高的查准率和查全率.