高光谱无损识别野生和种植黑枸杞

高光谱图像技术在农产品检测及识别方面有广阔的应用前景。野生黑枸杞经济效益显著,经常被种植黑枸杞冒充。提出一种利用高光谱图像对野生黑枸杞无损快速识别的方法。主要内容和结果如下：（1）共采集256份（野生、种植各128份）黑枸杞在900～1 700 nm范围的高光谱反射光谱,每份平均光谱作为此样品的光谱;（2）采用标准正态变换（SNV）对采集的光谱预处理;基于Kennard-Stone法,按照校正集和预测集比例为2∶1对样品划分,用连续投影算法（SPA）对光谱进行降维处理,提取特征波长30个;分别将全光谱和SPA 提取的30个特征波长作为模型输入,建立支持向量机（SVM）、极限学习机（ELM）和随机森林（RF）识别模型。（3）结果表明,在识别野生黑枸杞模型中,基于全光谱和SPA建立的SVM,ELM和RF模型校正集识别率均高于98.8%,基于全光谱和SPA建立的SVM,ELM和RF模型预测集识别率均高于97.7%。基于全光谱（FS）建立的三种识别模型略优于基于SPA建立的三种识别模型。但从简化模型方面,SPA提取的特征波常数仅为全光谱的11.8%,大大降低了模型运算量。三种模型中,基于随机森林模型无损识别野生黑枸杞效果最好,均达到100%。研究表明,利用高光谱图像技术结合分类模型可快速识别野生黑枸杞。     

基于全谱数据挖掘技术的土壤有机质高光谱预测建模研究

可见/近红外高光谱技术与建模方法是当前土壤近地传感器研究领域的重要方向,可应用于土壤养分信息的快速获取和农田作物的精确施肥管理。以浙江省水稻土为研究对象,利用以非线性模型为核心的数据挖掘技术,包括随机森林、支持向量机、人工神经网络等方法分别建立了不同建模集和验证集的原始光谱与有机质含量的估测模型。结果表明：研究比较的1∶1,3∶1和全部样本建模并全部验证的三种样本模式划分对建模的结果有一定的影响。相较于目前常用的偏最小二乘回归(PLSR)建模方法而言,非线性模型RF和SVM也取得了较好的建模精度,三种模式下其RDP值均大于1.4。特别是采用SVM建模方法所得模型具有很好的预测能力,模式二下其RDP值达到2.16。同时引入ANN方法改进建立的PLSR-ANN方法显著提高了PLSR的模型预测能力。     

宁夏盐池荒漠植物光谱特征分析及光谱识别

宁夏盐池县荒漠草地属于中温带干旱气候,由于过度利用出现不同程度的退化,退化指示种比重增大,造成不同荒漠草地群落组成差异也很大,如何区别不同荒漠草地植物,并据此对退化指示种进行动态监测是了解荒漠草地退化程度的关键。目前随机森林（RF）、支持向量机（SVM）与K-邻近（KNN）分类模型被广泛应用于森林植物和农作物的遥感分类,并取得了较好的分类识别效果,但针对草地尤其是荒漠草地植物的分类识别研究较少。因此使用ASD地物光谱仪于7月在宁夏盐池二步坑、冯记沟、高沙窝、麻黄山不同荒漠草地采集了32种植物作样本获得442条光谱进行光谱特征分析。筛选出7个植被指数：归一化植被指数705（NDVI₇₀₅）、绿通道植被指数（GNDVI）、光化学植被指数（PRI）、土壤调节植被指数（OSAVI）、可视化气压阻抗指数（VARI）、植被衰减指数（PSRI）和归一化水指数（NDWI）作为随机森林模型（RF）、支持向量机（SVM）模型、K-邻近（KNN）模型的原始变量,对32种荒漠草地植物进行分类识别,并通过分类模型精度的比较筛选较优模型。结果表明：(1）不同植物光谱反射率均符合绿色植物特征,但各植物原始光谱不同波段之间存在明显差异,植物原始光谱水分吸收波段差异明显,且有红边蓝移现象;(2）RF,SVM和KNN三个分类模型对32种植物的分类精度分别达到了0.98,0.94和0.98,识别效果较好,但3种分类模型均对白莲蒿与北芸香、虫实与甘草发生了误判;(3）随机森林模型重要性指标中NDWI与PRI为区分荒漠草地植物的关键指标,说明荒漠植物冠层水分与类胡萝卜素含量是影响荒漠草地植物光谱分类的重要因素。试验利用随机森林模型（RF）、支持向量机（SVM）与K-邻近（KNN）分类方法,建立了主要植物的分类模型。     

Classification of star/galaxy/QSO and star spectral types from LAMOST data release 5 with machine learning approaches

We use 343,747 sources from LAMOST DR5 to do star/galaxy/QSO classification with machine learning approaches. Specifically, the 312,767 spectral labeled stars (G, K, M, F, A) are used to do star classification. The photometry of u, g, r, i, z, J, and H are used as machine learning features. For star/galaxy/QSO classification, the k nearest neighbor algorithm (KNN), decision tree (DT), random forest (RF) and support vector machine (SVM) perform well. For star classification, the accuracy of RF and SVM classification are higher than the accuracy of KNN and DT. The area under receiver operating characteristic curves of the four models are approaching to 1. The accuracy, precision, recall, f_score, Matthews correlation coefficient are always greater than 0.5. The four models perform all right in predicting the nature of sources and the star label.     

Magnetic anisotropy and magnetization reversal of ultrathin iron films with in-plane magnetization on Si（111） substrates

The magnetic anisotropy and magnetization reversal of single crystal Fe films with thickness of 45 monolayer (ML) grown on Si(111) have been investigated by ferromagnetic resonance (FMR) and vibrating sample magnetometer (VSM). Owing to the significant modification of the energy surface in remanent state by slight misorientation from (111) plane and a uniaxial magnetic anisotropy, the azimuthal angular dependence of in-plane resonance field shows a six-fold symmetry with a weak uniaxial contribution, while the remanence of hysteresis loops displays a two-fold one. The competition between the first and second magnetocrystalline anisotropies may result in the switching of in-plane easy axis of the system. Combining the FMR and VSM measurements, the magnetization reversal mechanism has also been determined.     

Radio-frequency Mössbauer spectra of the “easy”-plane type magnetic system (FeBO3)

The theory of formation of the RF Mössbauer spectra for the “easy”-plane type magnetics (FeBO₃) and for various types of RF field polarization is presented. Experiments using both linearly and circularly polarized external RF fields were carried out at different temperatures. At room temperature the experimental spectra for both cases are well described by switching hyperfine (hf) field model. At temperatures close to the Neel temperature (335 K), the spectra in the oscillating and rotating RF field were obtained and their forms are described by models of switching and rotating hf field, respectively.     

Classification of bioaccumulative and non-bioaccumulative chemicals using statistical learning approaches

The present work aimed at developing in silico models allowing for a reliable prediction of bioaccumulative compounds and non-bioaccumulative compounds based on the definition of Bioconcentration Factor (BCF) using a diverse data set of 238 organic molecules. The partial least squares analysis (PLS), C4.5, support vector machine (SVM), and random forest (RF) algorithms were applied, and their performance classifying these compounds in terms of their quantitative structure-activity relationships (QSAR) was evaluated and verified with 5-fold cross-validation and an independent evaluation data set. The obtained results show that the overall prediction accuracies (Q) of the optimal PLS, C4.5, SVM and RF models are 84.5-87.7% for the internal cross-validation, with prediction accuracy (CO) of 86.3-91.1% in the external test sets, and C4.5 is slightly better than the three other methods which presents a Q of 87.7%, and a CO of 91.1% for the test sets. All these results prove the reliabilities of the in silico models, which should be valuable for the environmental risk assessment of the substances.     

Preparation,characterization and magneto-optical investigations of electrodeposited Co/Au films

Au/Co(4–8 ML)/Au single magnetic layers and Au(8 ML)/Co(4 ML)/Au(8 ML)/Co(8 ML)/Au bilayer were sequentially grown by electrodeposition on an Au(1 1 1) buffer layer electrodeposited on Si(1 1 1). The technique used in this work provides full control on the structure and the chemical composition of the different layers (no alloying) as well as on the chemistry at interfaces. scanning tunneling microscopy (STM) and atomic force microscopy (AFM) imaging and X-ray diffraction measurements show that atomically flat continuous Co(0 0 0 1) layers (4–8 ML) can be grown in epitaxy with the Au(1 1 1) substrate and that the 2 nm-thick spacer is also a continuous Au(1 1 1) layer. The Co ultrathin layers (4 and 8 ML) exhibit perpendicular magnetic anisotropy. The lateral magnetic homogeneity and magnetization reversal process have been investigated by scanning magneto-optical Kerr effect (MOKE) magnetometry and global Kerr microscopy. The correlation between magnetization switching behaviour in each layer of the Co-bilayer stack has been evidenced from in-depth sensitive MOKE measurements and microscopy. The strong coupling observed between the two Co layers is attributed to magnetostatic interaction at domain wall boundaries.     

Magnetotransport and magnetic properties of InAs/Mn digital alloys

We have investigated the magnetic and magneto-transport properties of a systematic sequence of five InAs/Mn digital alloys grown by a combination of molecular beam epitaxy and atomic layer epitaxy. The samples consist of 30 periods of Mn fractional monolayers (ML) (0.17–0.5 ML) separated by 14 ML thick InAs spacer layers in a superlattice configuration. Four samples show n-type electrical conduction while the fifth (0.25 ML Mn) is p-type. Squid magnetization measurements performed on these samples show remnant magnetization above room temperature, which is apparently related to a second phase.     

Perpendicular magnetic anisotropy of ultrathin FeCo alloy films on Pd(0 0 1) surface: First principles study

Using the full potential linearized augmented plane wave (FLAPW) method, thickness dependent magnetic anisotropy of ultrathin FeCo alloy films in the range of 1 monolayer (ML) to 5 ML coverage on Pd(0 0 1) surface has been explored. We have found that the FeCo alloy films have close to half metallic state and well-known surface enhancement in thin film magnetism is observed in Fe atom, whereas the Co has rather stable magnetic moment. However, the largest magnetic moment in Fe and Co is found at 1 ML thickness. Interestingly, it has been observed that the interface magnetic moments of Fe and Co are almost the same as those of surface elements. The similar trend exists in orbital magnetic moment. This indicates that the strong hybridization between interface FeCo alloy and Pd gives rise to the large magnetic moment. Theoretically calculated magnetic anisotropy shows that the 1 ML FeCo alloy has in-plane magnetization, but the spin reorientation transition (SRT) from in-plane to perpendicular magnetization is observed above 2 ML thickness with huge magnetic anisotropy energy. The maximum magnetic anisotropy energy for perpendicular magnetization is as large as 0.3 meV/atom at 3 ML film thickness with saturation magnetization of . Besides, the calculated X-ray magnetic circular dichroism (XMCD) has been presented.     

Deep learning assisted time-varying channel estimation in multi-user mmWave hybrid MIMO systems

Hybrid analog/digital multiple input multiple output (MIMO) system is proposed to mitigate the challenges of millimeter wave (mmWave) communication. This architecture enables utilizing the large array gain with reasonable power consumption. However, new methods are required for the channel estimation problem of hybrid architecture-based systems due to the fewer number of radio frequency (RF) chains than antenna elements. Leveraging the sparse nature of the mmWave channels, compressed sensing (CS)-based channel estimation methods are proposed. Recently, machine learning (ML)-aided methods have been investigated to improve the channel estimation performance. Additionally, the Doppler effect should be considered for the high mobility scenarios, and we deal with the time-varying channel model. Therefore, in this article, we consider the scenario of time-varying channels for a multi-user mmWave hybrid MIMO system. By proposing a Deep Neural Network (DNN) and defining the inputs and outputs, we introduce a novel algorithm called Deep Learning Assisted Angle Estimation (DLA-AE) for improving the estimation of the Angles of Departure/Arrival (AoDs/AoAs) of the channel paths. In addition, we suggest Linear Phase Interpolation (LPI) to acquire the path gains for the data transmission instants. Simulation results show that utilizing the proposed DLA-AE and LPI methods enhance the time-varying channel estimation accuracy with low computational complexity.     

基于SVM与RF的苹果树冠LAI高光谱估测

叶面积指数(leaf area index,LAI)是反映作物群体大小的较好的动态指标。运用高光谱技术快速、无损地估测苹果树冠叶面积指数,为监测苹果树长势和估产提供参考。以盛果期红富士苹果树为研究对象,采用ASD地物光谱仪和LAI-2200冠层分析仪,在山东省烟台栖霞研究区,连续2年测量了30个果园90棵苹果树冠层光谱反射率及LAI值;通过相关性分析方法构建并筛选出了最优的植被指数;利用支持向量机(support vector machine, SVM)与随机森林(random forests, RF)多元回归分析方法构建了LAI估测模型。新建的GNDVI₅₂₇,NDVI₆₇₆,RVI₆₈₂,FD-NVI₆₅₆和GRVI₅₁₇五个植被指数及前人建立的两个植被指数NDVI₆₇₀和NDVI₇₀₅与LAI的相关性都达到了极显著水平;建立的RF回归模型中,校正集决定系数C-R2和验证集决定系数V-R2为0.920,0.889,分别比SVM回归模型提高了0.045和0.033,校正集均方根误差C-RMSE、验证集均方根误差V-RMSE为0.249,0.236,分别比SVM回归模型降低了0.054和0.058, 校正集相对分析误C-RPD、验证集相对分析误V-RPD达到了3.363和2.520,分别比SVM回归模型提高了0.598和0.262,校正集及验证集的实测值与预测值散点图趋势线的斜率C-S和V-S都接近于1,RF回归模型的估测效果优于SVM。RF多元回归模型适合盛果期红富士苹果树LAI的估测。     

Machine Learning to Instruct Single Crystal Growth by Flux Method

Growth of high-quality single crystals is of great significance for research of condensed matter physics. The exploration of suitable growing conditions for single crystals is expensive and time-consuming, especially for ternary compounds because of the lack of ternary phase diagram. Here we use machine learning(ML) trained on our experimental data to predict and instruct the growth. Four kinds of ML methods, including support vector machine(SVM), decision tree, random forest and gradient boosting decision tree, are adopted. The SVM method is relatively stable and works well, with an accuracy of 81% in predicting experimental results. By comparison,the accuracy of laboratory reaches 36%. The decision tree model is also used to reveal which features will take critical roles in growing processes.     

Deep learning-based channel estimation using Gaussian mixture distribution and expectation maximum algorithm

In massive multiple-input multiple-output (MIMO), it is much challenging to obtain accurate channel state information (CSI) after radio frequency (RF) chain reduction due to the high dimensions. With the fast development of machine learning(ML), it is widely acknowledged that ML is an effective method to deal with channel models which are typically unknown and hard to approximate. In this paper, we use the low complexity vector approximate messaging passing (VAMP) algorithm for channel estimation, combined with a deep learning framework for soft threshold shrinkage function training. Furthermore, in order to improve the estimation accuracy of the algorithm for massive MIMO channels, an optimized threshold function is proposed. This function is based on Gaussian mixture (GM) distribution modeling, and the expectation maximum Algorithm (EM Algorithm) is used to recover the channel information in beamspace. This contraction function and deep neural network are improved on the vector approximate messaging algorithm to form a high-precision channel estimation algorithm. Simulation results validate the effectiveness of the proposed network.     

Anatomy of field effects on magnetization dynamics and spin transfer noise

Spin transfer-related phenomena in nanomagnets have attracted extensive studies. In this paper we shall focus on analysis of individual and combined effects of the external, anisotropy, and demagnetization fields on magnetization dynamics and spin transfer noise. It is found that individual roles of the external, anisotropy, and demagnetization fields, as well as the combined roles of external plus anisotropy fields and anisotropy plus demagnetization fields, do not change the behavior of current induced magnetization switching. Such magnetization reversal procedures are of low noise. Our dynamics and power spectral density calculations show that it is the demagnetization field that plays a major role in inducing spin transfer noise: the demagnetization field itself or in combination with the anisotropy field will result in wave-like switching; moreover, the demagnetization field, together with the external field (not too small), will lead to precession and hence the system would be in noisy states. Our modeling work for an elliptical Py alloy is qualitatively consistent with Cornell's experiment and simulation [Science 307 (2005) 228].     

Tailoring the optimal control cost function to a desired output: application to minimizing phase errors in short broadband excitation pulses

The de facto standard cost function has been used heretofore to characterize the performance of pulses designed using optimal control theory. The freedom to choose new, creative quality factors designed for specific purposes is demonstrated. While the methodology has more general applicability, its utility is illustrated by comparison to a consistently chosen example--broadband excitation. The resulting pulses are limited to the same maximum RF amplitude used previously and tolerate the same variation in RF homogeneity deemed relevant for standard high-resolution NMR probes. Design criteria are unchanged: transformation of I(z)--> I(x) over resonance offsets of +/-20 kHz and RF variability of +/-5%, with a peak RF amplitude equal to 17.5 kHz. However, the new cost effectively trades a small increase in residual z magnetization for improved phase in the transverse plane. Compared to previous broadband excitation by optimized pulses (BEBOP), significantly shorter pulses are achievable, with only marginally reduced performance. Simulations transform I(z) to greater than 0.98 I(x), with phase deviations of the final magnetization less than 2 degrees, over the targeted ranges of resonance offset and RF variability. Experimental performance is in excellent agreement with the simulations.     

RF magnetron sputtering induced the perpendicular magnetic anisotropy modification in Pt/Co based multilayers

We demonstrate that radio frequency(RF)magnetron sputtering technique can modify the perpendicular magnetic anisotropy(PMA)of Pt/Co/normal metal(NM)thin films.Influence of ion irradiation during RF magnetron sputtering should not be neglected and it can weaken PMA of the deposited magnetic films.The magnitude of this influence can be controlled by tuning RF magnetron sputtering deposition conditions and the upper NM layer thickness.According to the stopping and range of ions in matter(SRIM)simulation results,defects such as displacement atoms and vacancies in the deposited film will increase after the RF magnetron sputtering,which can account for the weakness of PMA.The amplitude changes of the Hall resistance and the threshold current intensity of spin orbit torque(SOT)induced magnetization switching also can be modified.Our study could be useful for controlling magnetic properties of PMA films and designing new type of SOT-based spintronic devices.     

A deep neural network surrogate modeling benchmark for temperature field prediction of heat source layout

The thermal issue is of great importance during the layout design of heat source components in systems engineering, especially for high functional-density products. Thermal analysis requires complex simulation, which leads to an unaffordable computational burden to layout optimization as it iteratively evaluates different schemes. Surrogate modeling is an effective method for alleviating computation complexity. However, the temperature field prediction(TFP) with complex heat source layout(HSL) input is an ultra-high dimensional nonlinear regression problem, which brings great difficulty to traditional regression models. The deep neural network(DNN) regression method is a feasible way for its good approximation performance. However, it faces great challenges in data preparation for sample diversity and uniformity in the layout space with physical constraints and proper DNN model selection and training for good generality, which necessitates the efforts of layout designers and DNN experts. To advance this cross-domain research, this paper proposes a DNN-based HSL-TFP surrogate modeling task benchmark. With consideration for engineering applicability, sample generation, dataset evaluation, DNN model, and surrogate performance metrics are thoroughly investigated. Experiments are conducted with ten representative state-of-the-art DNN models. A detailed discussion on baseline results is provided, and future prospects are analyzed for DNN-based HSL-TFP tasks.