光谱数据解析中的变量筛选方法

如何从海量或高维数据中“提纯”出有用的信息,这是当前数据分析面临的一个巨大的挑战,也是当前研究的一个热点。变量筛选技术能够从众多、复杂的量测数据中提取出特征信息变量,达到简化多元模型乃至提高模型预测性能等目的。在光谱分析中,来自噪声等诸多因素的影响,量测数据会不可避免地包含干扰和无关信息变量,以及变量间存在的多重共线性,这些都会影响模型的稳健性和预测能力。近年来变量（波长）筛选方法在光谱解析领域的研究与应用中取得了较大的进展。结合国内外相关研究文献和作者的研究体会,不仅仅综述了近红外光谱,还综述了中红外光谱、拉曼光谱等众多筛选变量的方法的提出、特点、发展、类别、比较和近五年来在不同领域的应用进展。其中,评价变量重要性的参数及其标准或阈值的选择、搜索变量的策略和途径是变量筛选方法的关键。而且每种方法都具有各自的优势和局限性,实际使用中要根据方法自身特点结合目标体系的特征选择合适的方法。重点内容：（1）对比了光谱数据分析中常用的波长筛选和波段筛选方法;（2）对比了基于PLS模型参数的不同变量筛选方法的原理和特点;（3）根据搜索和筛选变量策略的不同将变量筛选方法进行分类评述。最后,围绕在解析实际复杂体系中变量筛选方法出现的过拟合、不稳定等问题进行了讨论并提出相应的解决措施,同时对变量筛选方法的研究趋势、发展前景和应用方向进行了展望。其中,新的评价变量重要性的判据和搜索变量的策略等工作仍需要展开深入地研究。期望本综述能够对光谱变量筛选的后续研究及应用起到积极的推动作用。     

EBCMOS相机盲元检测方法的研究

电子轰击型(electronbombarded,EB)CMOS传感器工艺缺陷导致图像出现盲元,对图像质量影响较大.提出了一种基于多幅图像交叉计算的盲元检测新算法.对不同平均亮度的无目标图像,通过设置不同阈值,对图像进行二值化处理;再对多幅二值化图像按"亮点"位置进行交叉"相与"计算,生成盲元标记模板图,最后按盲元位置进...     

Gold Nanoparticles: Production,Reshaping, and Thermal Charging

Gold nanoparticles are of great interest for various nanoelectronic applications, e.g., for making single electron transistors or very fine leads to molecular size entities. For this and other applications, it is important that all particles have controllable size and shape. In this paper, we describe the production of size-selected gold aerosol particles in the 20nm range made by evaporation in a high-temperature tube furnace and subsequent size selection. To obtain spherical particles, it was necessary to reshape the particles at high temperature, which was investigated for temperatures between 25°C and 1200°C. High-resolution transmission electron microscopy showed that the degree of crystallinity became higher for higher reshaping temperature. During reshaping at high temperature, an anomalous charging behavior was discovered, whereby negatively as well as positively charged particles became multiply negatively charged. Possible mechanisms for explaining this thermally activated phenomenon are discussed.     

基于K-L散度与光谱可分性距离的波段选择算法

波段选择是高光谱降维的常用手段,文中从波段选择应遵循的3个原则出发设计了一种基于信息散度与光谱可分性距离的波段选择算法。将高光谱数据中每个波段的光谱分量看作一个一维向量,使用K-L散度表示其相互之间的信息量,选出信息量大且相似性最小的波段组合;根据每个波段中不同地物光谱可分性距离的计算,得到可分性较大的波段组合;将两组波段组合取交集,即得到最优组合波段。为了验证算法的有效性,将选出的最佳3个波段进行伪彩色合成,对其进行光谱角制图分类,分类精度达到922%,Kappa系数为088.     

Two-stage blind deconvolution scheme using useful priors

Hand shake blurry image is a common phenomenon in our daily life. In this paper, a novel blind deconvolution scheme is proposed to recover a single hand shake blurry image. The algorithm is subdivided into two main stages, kernel estimation stage and non-blind deconvolution stage. In the kernel estimation stage, we propose a cost function taking a selected map into consideration. In the non-blind decovolution stage, another cost function is designed using image derivatives prior. We also present an adaptive kernel size selection method instead of traditional manual selection. Extensive experiments on real world blurry images are conducted to demonstrate the performance of our algorithm.     

Classification of flaw severity using pattern recognition for guided wave-based structural health monitoring

In this paper, the authors present a formal classification routine to characterize flaw severity in an aircraft-grade aluminum plate using Lamb waves. A rounded rectangle flat-bottom hole is incrementally introduced into the plate, and at each depth multi-mode Lamb wave signals are collected to study the changes in received signal due to mode conversion and scattering from the flaw. Lamb wave tomography reconstructions are used to locate and size the flaw at each depth, however information about the severity of the flaw is obscured when the flaw becomes severe enough that scattering effects dominate. The dynamic wavelet fingerprint is then used to extract features from the raw Lamb wave signals, and supervised pattern classification techniques are used to identify flaw severity with up to 80.7% accuracy for a training set and up to 51.7% accuracy on a series of validation data sets extracted from independent plate samples.     

从随机到取向——分子转动动力学的超快调控

分子取向物理是研究如何操控分子空间排列方向以及如何制备具有特定方向的分子。近十多年来,分子取向物理学取得巨大进展,人们通过各种技术对分子取向物理进行研究,使人们对分子取向的认识达到一个全新的高度。文章首先介绍了分子的量子态及其分布,转动量子态选择方法和分子的转动动力学,然后概述通过制备和调控转动态的相干布居来控制分子取向的方法。实验和理论表明,分子的不同转动态在相同激光条件作用下产生的转动波包不一样,从而导致分子的准直和取向程度随时间演化完全不一样。通过选择合适的转动量子态能够很好地提高分子取向程度。取向分子作为相互作用靶,为原子分子动力学、精密测量物理、立体化学反应和表面碰撞过程等提供了更加丰富多彩的研究内容。     

MIMO Self-Heterodyne OFDM Using Band Selection Technique

The 5G technology is a promising technology to cope with the increasing demand for higher data rate and quality of service. In this paper, two proposed techniques are implemented for multiple input multiple output (MIMO) self-heterodyne OFDM system to enhance data rate and minimize the bit error rate (BER). In both of the two proposed techniques, Band Selection (BS) approach is used, once with Space Time Block Coded (STBC) for the first proposed technique (BS- STBC), and once again with Frequency Space Time Block Coded (FSTBC) for the second proposed technique (BS-FSTBC). The use of the BS in the proposed techniques helps to choose the sub-band with better subchannels gains for sending the information and consequently, minimize the BER. Moreover, the use of the FSTBC instead of STBC helps to use the spectral efficiently and hence increase data rate. The simulation results show that the proposed techniques BS-STBC and BS-FSTBC, for the MIMO self-heterodyne OFDM system, provide a great enhancement in the BER performance when compared to the conventional techniques. Moreover, the simulation results show that the first proposed technique BS-FSTBC outperform the second propose technique BS-STBC in term of the BER performance.     

Causal Geometry

Information geometry has offered a way to formally study the efficacy of scientific models by quantifying the impact of model parameters on the predicted effects. However, there has been little formal investigation of causation in this framework, despite causal models being a fundamental part of science and explanation. Here, we introduce causal geometry, which formalizes not only how outcomes are impacted by parameters, but also how the parameters of a model can be intervened upon. Therefore, we introduce a geometric version of “effective information”—a known measure of the informativeness of a causal relationship. We show that it is given by the matching between the space of effects and the space of interventions, in the form of their geometric congruence. Therefore, given a fixed intervention capability, an effective causal model is one that is well matched to those interventions. This is a consequence of “causal emergence,” wherein macroscopic causal relationships may carry more information than “fundamental” microscopic ones. We thus argue that a coarse-grained model may, paradoxically, be more informative than the microscopic one, especially when it better matches the scale of accessible interventions—as we illustrate on toy examples.     

A Two-Stage Hybrid Default Discriminant Model Based on Deep Forest

Background: the credit scoring model is an effective tool for banks and other financial institutions to distinguish potential default borrowers. The credit scoring model represented by machine learning methods such as deep learning performs well in terms of the accuracy of default discrimination, but the model itself also has many shortcomings such as many hyperparameters and large dependence on big data. There is still a lot of room to improve its interpretability and robustness. Methods: the deep forest or multi-Grained Cascade Forest (gcForest) is a decision tree depth model based on the random forest algorithm. Using multidimensional scanning and cascading processing, gcForest can effectively identify and process high-dimensional feature information. At the same time, gcForest has fewer hyperparameters and has strong robustness. So, this paper constructs a two-stage hybrid default discrimination model based on multiple feature selection methods and gcForest algorithm, and at the same time, it optimizes the parameters for the lowest type II error as the first principle, and the highest AUC and accuracy as the second and third principles. GcForest can not only reflect the advantages of traditional statistical models in terms of interpretability and robustness but also take into account the advantages of deep learning models in terms of accuracy. Results: the validity of the hybrid default discrimination model is verified by three real open credit data sets of Australian, Japanese, and German in the UCI database. Conclusions: the performance of the gcForest is better than the current popular single classifiers such as ANN, and the common ensemble classifiers such as LightGBM, and CNNs in type II error, AUC, and accuracy. Besides, in comparison with other similar research results, the robustness and effectiveness of this model are further verified.