BASIC THEORY AND APPLICATIONS OF WELDING ARC SPECTRAL INFORMATION

Welding arc spectral information is a rising welding information source. In some occasion, it can reflect many physical phenomena of welding process and solve many problems that cannot be done with arc electric information, acoustic information and other arc information. It is of important significance in developing automatic control technique of welding process and other similar process. Many years study work on welding arc spectral information of the anthor are discussed from three aspects of theory, method and application. Basic theory, view and testing methods of welding arc spectral information has been put forward. In application aspects, many applied examples, for example, monitoring of harmful gases in arc （such as hydrogen and nitrogen） with the method of welding arc spectral information; welding arc spectral imaging of the welding pool which is used in automatic seam tracking; controlling of welding droplet transfer with welding arc spectral information and so on, are introduced. Especially, the successful application in real time controlling of welding droplet transfer in pulsed GMAW is introduced too. These application examples show that the welding arc spectral information has great applied significance and development potentialities. These .content will play an important role in applying and spreading welding arc spectral informarion technology.

Color Restoration Method Based on Spectral Information Using Normalized Cut

This paper proposes a novel method for color restoration that can effectively apply accurate color based on spectral information to a segmented image using the normalized cut technique. Using the proposed method, we can obtain a digital still camera image and spectral information in different environments. Also, it is not necessary to estimate reflectance spectra using a spectral database such as other methods. The synthesized images are accurate and high resolution. The proposed method effectively works in making digital archive contents. Some experimental results are demonstrated in this paper.

BASIC THEORY AND METHOD OF WELDING ARC SPECTRAL INFORMATION

Arc spectral information is a rising information source which can solve manyproblems that can not be done with arc electric information and other arc information. It is ofimportant significance to develop automatic control technique of welding process. The basic theoryand methods on it play an important role in expounding and applying arc spectral information. Usingconcerned equation in plasma physics and spectrum theory, a system of equations including 12equations which serve as basic theory of arc spectral information is set up. Through analyzing ofthe 12 equations, a basic view that arc spectral information is the reflection of arc state andstate variation, and is the most abundant information resource reflecting welding arc process isdrawn. Furthermore, based on the basic theory, the basic methods of test and control of arc spectralinformation and points out some applications of it are discussesed.

Solution of Multiple-Point Statistics to Extracting Information from Remotely Sensed Imagery

Two phenomena of similar objects with different spectra and different objects with similar spectrum often result in the difficulty of separation and identification of all types of geographical objects only using spectral information. Therefore, there is a need to incorporate spatial structural and spatial association properties of the surfaces of objects into image processing to improve the accuracy of classification of remotely sensed imagery. In the current article, a new method is proposed on the basis of the principle of multiple-point statistics for combining spectral information and spatial information for image classification. The method was validated by applying to a case study on road extraction based on Landsat TM taken over the Chinese Yellow River delta on August 8, 1999. The classification results have shown that this new method provides overall better results than the traditional methods such as maximum likelihood classifier （MLC）.

Study on Remote Sensing Information Extraction Technology for the Impervious Surface of Erhai Basin

[Objective] To study the remote sensing information extraction technology for the impervious surface of Erhai basin with the aim to develop dynamic simulation platform for the formation of water pollution. [Method] Linear spectral separation technology was used to achieve Vd-S model solution, extracting remote sensing in- formation of the impervious surface of Erhai basin from the TM data of Landsat5 in 2009. The linear combination of 4 kinds of endmember spectra, namely vegetation, high anti-illumination, low anti-illumination and bare soil, were used to simulate the TM spectral characteristics, and its distribution and spatial characteristics were ana- lyzed. [Result] Middle-resolution image is suitable for the basin-scaled impervious surface extraction with reliable results and satisfactory accuracy. [Conclusion] This study provided basis for deciding the relationship between the regulation strategy on the non-point source pollution of Erhai Lake, coordinated economic development and environmental protection.

Investigation on the Indeterminate Information of Rock Joint Roughness through a Neutrosophic Number Approach

To better estimate the rock joint shear strength,accurately determining the rock joint roughness coefficient(JRC)is the first step faced by researchers and engineers.However,there are incomplete,imprecise,and indeterminate problems during the process of calculating the JRC.This paper proposed to investigate the indeterminate information of rock joint roughness through a neutrosophic number approach and,based on this information,reported a method to capture the incomplete,uncertain,and imprecise information of the JRC in uncertain environments.The uncertainties in the JRC determination were investigated by the regression correlations based on commonly used statistical parameters,which demonstrated the drawbacks of traditional JRC regression correlations in handling the indeterminate information of the JRC.Moreover,the commonly used statistical parameters cannot reflect the roughness contribution differences of the asperities with various scales,which induces additional indeterminate information.A method based on the neutrosophic number(NN)and spectral analysis was proposed to capture the indeterminate information of the JRC.The proposed method was then applied to determine the JRC values for sandstone joint samples collected from a rock landslide.The comparison between the JRC results obtained by the proposed method and experimental results validated the effectiveness of the NN.Additionally,comparisons made between the spectral analysis and common statistical parameters based on the NN also demonstrated the advantage of spectral analysis.Thus,the NN and spectral analysis combined can effectively handle the indeterminate information in the rock joint roughness.

Intervention against information diffusion in static and temporal coupling networks

Information diffusion in complex networks has become quite an active research topic.As an important part of this field,intervention against information diffusion processes is attracting ever-increasing attention from network and control engineers.In particular,it is urgent to design intervention schemes for the coevolutionary dynamics between information diffusion processes and coupled networks.For this purpose,we comprehensively study the problem of information diffusion intervention over static and temporal coupling networks.First,individual interactions are described by a modified activitydriven network(ADN)model.Then,we establish a novel node-based susceptible-infected-recovered-susceptible(SIRS)model to characterize the information diffusion dynamics.On these bases,three synergetic intervention strategies are formulated.Second,we derive the critical threshold of the controlled-SIRS system via stability analysis.Accordingly,we exploit a spectral optimization scheme to minimize the outbreak risk or the required budget.Third,we develop an optimal control scheme of dynamically allocating resources to minimize both system loss and intervention expense,in which the optimal intervention inputs are obtained through optimal control theory and a forward-backward sweep algorithm.Finally,extensive simulation results validate the accuracy of theoretical derivation and the performance of our proposed intervention schemes.

Intelligent Biometric Information Management

We advance here a novel methodology for robust intelligent biometric information management with inferences and predictions made using randomness and complexity concepts. Intelligence refers to learning, adap- tation, and functionality, and robustness refers to the ability to handle incomplete and/or corrupt adversarial information, on one side, and image and or device variability, on the other side. The proposed methodology is model-free and non-parametric. It draws support from discriminative methods using likelihood ratios to link at the conceptual level biometrics and forensics. It further links, at the modeling and implementation level, the Bayesian framework, statistical learning theory (SLT) using transduction and semi-supervised lea- rning, and Information Theory (IY) using mutual information. The key concepts supporting the proposed methodology are a) local estimation to facilitate learning and prediction using both labeled and unlabeled data;b) similarity metrics using regularity of patterns, randomness deficiency, and Kolmogorov complexity (similar to MDL) using strangeness/typicality and ranking p-values;and c) the Cover – Hart theorem on the asymptotical performance of k-nearest neighbors approaching the optimal Bayes error. Several topics on biometric inference and prediction related to 1) multi-level and multi-layer data fusion including quality and multi-modal biometrics;2) score normalization and revision theory;3) face selection and tracking;and 4) identity management, are described here using an integrated approach that includes transduction and boosting for ranking and sequential fusion/aggregation, respectively, on one side, and active learning and change/ outlier/intrusion detection realized using information gain and martingale, respectively, on the other side. The methodology proposed can be mapped to additional types of information beyond biometrics.

Stacked spectral feature space patch: An advanced spectral representation for precise crop classification based on convolutional neural network

Spectral and spatial features in remotely sensed data play an irreplaceable role in classifying crop types for precision agriculture. Despite the thriving establishment of the handcrafted features, designing or selecting such features valid for specific crop types requires prior knowledge and thus remains an open challenge. Convolutional neural networks(CNNs) can effectively overcome this issue with their advanced ability to generate high-level features automatically but are still inadequate in mining spectral features compared to mining spatial features. This study proposed an enhanced spectral feature called Stacked Spectral Feature Space Patch(SSFSP) for CNN-based crop classification. SSFSP is a stack of twodimensional(2 D) gridded spectral feature images that record various crop types’ spatial and intensity distribution characteristics in a 2 D feature space consisting of two spectral bands. SSFSP can be input into2 D-CNNs to support the simultaneous mining of spectral and spatial features, as the spectral features are successfully converted to 2 D images that can be processed by CNN. We tested the performance of SSFSP by using it as the input to seven CNN models and one multilayer perceptron model for crop type classification compared to using conventional spectral features as input. Using high spatial resolution hyperspectral datasets at three sites, the comparative study demonstrated that SSFSP outperforms conventional spectral features regarding classification accuracy, robustness, and training efficiency. The theoretical analysis summarizes three reasons for its excellent performance. First, SSFSP mines the spectral interrelationship with feature generality, which reduces the required number of training samples.Second, the intra-class variance can be largely reduced by grid partitioning. Third, SSFSP is a highly sparse feature, which reduces the dependence on the CNN model structure and enables early and fast convergence in model training. In conclusion, SSFSP has great potential for practical crop classification in precision agriculture.

Piecewise spectrally band-pass for compressive coded aperture spectral imaging

Coded aperture snapshot spectral imaging(CASSI) has been discussed in recent years. It has the remarkable advantages of high optical throughput, snapshot imaging, etc. The entire spatial-spectral data-cube can be reconstructed with just a single two-dimensional(2D) compressive sensing measurement. On the other hand, for less spectrally sparse scenes,the insufficiency of sparse sampling and aliasing in spatial-spectral images reduce the accuracy of reconstructed threedimensional(3D) spectral cube. To solve this problem, this paper extends the improved CASSI. A band-pass filter array is mounted on the coded mask, and then the first image plane is divided into some continuous spectral sub-band areas. The entire 3D spectral cube could be captured by the relative movement between the object and the instrument. The principle analysis and imaging simulation are presented. Compared with peak signal-to-noise ratio(PSNR) and the information entropy of the reconstructed images at different numbers of spectral sub-band areas, the reconstructed 3D spectral cube reveals an observable improvement in the reconstruction fidelity, with an increase in the number of the sub-bands and a simultaneous decrease in the number of spectral channels of each sub-band.

Quantum-Classical Algorithm for an Instantaneous Spectral Analysis of Signals:A Complement to Fourier Theory

A quantum time-dependent spectrum analysis, or simply, quantum spectral analysis (QSA) is presented in this work, and it’s based on Schrödinger’s equation. In the classical world, it is named frequency in time (FIT), which is used here as a complement of the traditional frequency-dependent spectral analysis based on Fourier theory. Besides, FIT is a metric which assesses the impact of the flanks of a signal on its frequency spectrum, not taken into account by Fourier theory and lets alone in real time. Even more, and unlike all derived tools from Fourier Theory (i.e., continuous, discrete, fast, short-time, fractional and quantum Fourier Transform, as well as, Gabor) FIT has the following advantages, among others: 1) compact support with excellent energy output treatment, 2) low computational cost, O(N) for signals and O(N2) for images, 3) it does not have phase uncertainties (i.e., indeterminate phase for a magnitude = 0) as in the case of Discrete and Fast Fourier Transform (DFT, FFT, respectively). Finally, we can apply QSA to a quantum signal, that is, to a qubit stream in order to analyze it spectrally.

Toward Circumventing Collinearity Effect in Nonlinear Spectral Mixture Analysis by Using a Spectral Shape Measure

Nonlinear spectral mixture analysis (NSMA) is a widely used unmixing algorithm. It can fit the mixed spectra adequately, but collinearity effect among true and virtual endmembers will decrease the retrieval accuracies of endmember fractions. Use of linear spectral mixture analysis (LSMA) can effectively reduce the degree of collinearity in the NSMA. However, the inadequate modeling of mixed spectra in the LSMA will also yield retrieval errors, especially for the cases where the multiple scattering is not ignorable. In this study, a generalized spectral unmixing scheme based on a spectral shape measure, i.e. spectral information divergence (SID), was applied to overcome the limitations of the conventional NSMA and LSMA. Two simulation experiments were undertaken to test the performances of the SID, LSMA and NSMA in the mixture cases of treesoil, tree-concrete and tree-grass. Results demonstrated that the SID yielded higher accuracies than the LSMA for almost all the mixture cases in this study. On the other hand, performances of the SID method were comparable with the NSMA for the tree-soil and tree-grass mixture cases, but significantly better than the NSMA for the tree-concrete mixture case. All the results indicate that the SID method is fairly effective to circumvent collinearity effect within the NSMA, and compensate the inadequate modeling of mixed spectra within the LSMA.

植被覆盖区老旧矿山边部遥感蚀变信息提取技术探讨

植被覆盖区基岩裸露少,围岩蚀变信息微弱.为在老旧矿山边部开展找矿工作,通过不同方法实验,最后采用植被抑制+岩土体光谱测试组合技术完成蚀变信息提取.该方法首先运用强迫不变植被抑制技术降低植被干扰,其次采集已有矿山蚀变围岩及蚀变带岩土体光谱信息并建立研究区光谱样本库,最后采用CART决策树分类技术完成蚀变信息提取.实验证明,基于植被抑制技术的CART决策树分类蚀变信息提取方法在植被覆盖区老旧矿山边部遥感蚀变信息提取效果较好.

太赫兹时域光谱成像增强算法

低分辨率是扫描太赫兹时域光谱(THz-TDS)系统图像的一个重要缺点.本文提出了一种针对太赫兹时域光谱图像的超分辨率增强算法,该算法结合了多层感知机(MLP)和超分辨率卷积神经网络(SRCNN),创建了一个复合网络.本文算法针对太赫兹光谱图像的特点,通过引入新的目标函数,避免了传统机器学习算法需要采集或模拟生成大量训练集的弊端,实现了训练图像即目标图像的单幅光谱图像增强.为了实现这一目标,本文算法的基本原理是,让试件产生一个刚体位移,利用THz-TDS系统采集位移前后两幅三维(1个时间维,2个空间维)光谱图像作为输入数据.机器学习网络包括两部分:首先,利用一个MLP网络实现三维光谱图像到二维光强图像的转化;其次,采用传统针对二维图像的SRCNN网络获取一幅高分辨率图像,对位移前后图像处理后计算得到新的高分辨率图像的位移场,并将位移场方差作为目标函数,再通过机器学习算法,优化网络中的成像参数,实现太赫兹光谱图像的分辨率增强.典型验证性实验最终得到的峰值信噪比为42.65 dB,结构相似度为0.816,均比其他现有方法高,表明本文算法能获得良好的图像增强.

基于鞍山式铁矿成像光谱的融合算法研究

铁矿资源是我国经济发展和社会进步的物质基础。在铁矿开采过程中,快速精准地确定铁矿品位,对矿山开采决策及经济效益具有重要影响。高光谱成像技术具有影像覆盖范围广、精度高等优势,已广泛应用于矿石分类及成分反演等领域。然而目前高光谱成像传感器的波段范围主要为可见短近红外(Vis-SWIR)和近红外(NIR)两类,且两类数据多为独立获取,缺乏连续性,采用单一数据所建模型的精度往往偏低。因此融合多传感器所获光谱数据,可有效解决单一传感器波段范围小、包含目标特征波段少等问题,提高基于高光谱成像技术的铁矿品位反演精度。使用Pika L与Pika NIR-320高光谱成像仪,分别在Vis-SWIR与NIR两个波段范围内采集鞍山式铁矿的成像光谱数据,提出了基于互信息(MI)的光谱串联融合方法,该方法首先对两组光谱数据进行预处理,然后对处理后的数据进行互信息计算以此对光谱数据进行串联融合。最后分别以Vis-SWIR、NIR以及基于不同波段串联融合的光谱数据为数据源,建立RBF神经网络品位反演模型,并以融合前后光谱数据所建模型的准确性与精度为融合算法有效性的判别指标。结果表明,光谱数据串联融合后所建模型的准确性与精度高于单独使用Vis-SWIR、NIR光谱数据所建模型的准确性与精度。与基于其余波段串联融合的光谱数据相比,在基于互信息计算得出的959.89nm处串联融合后光谱数据所建模型的准确性与精度最高,R2为0.88,RPD为2.97,RMSE为4.464,MAE为3.32。该研究针对多传感器光谱融合提出了一种新思路,对成像光谱技术应用于铁矿品位反演具有现实指导意义。

基于GF-5卫星的西藏珠勒—芒拉地区矿物蚀变信息提取及找矿前景分析

【研究目的】近年来,遥感在地质调查和矿产勘查领域取得了广泛的应用,基于多光谱遥感数据的蚀变矿物填图为地质找矿工作提供了重要技术支撑,然而基于国产高光谱遥感数据在此领域的研究却为数不多。高分五号(GF-5)较小的波谱间隔提供了相比于多光谱更为丰富的目标地物波谱信息,为矿物的精细识别提供了良好的数据源。本文主要基于GF-5开展西藏革吉南地区的矿物蚀变信息提取,同时结合Landsat-8、ASTER多光谱数据提取结果叠加对比,综合野外调查验证,进一步深化遥感在地质矿产资源调查领域的应用。【研究方法】基于多光谱数据建立了不同类别蚀变矿物的光谱指数模型,在GF-5数据蚀变信息提取方面,摒弃了传统的光谱角匹配等方法,提出了基于决策树分类辅助混合调谐匹配滤波技术进行矿化蚀变信息的提取方法,最后综合区域构造、蚀变信息提取结果等要素,圈定成矿有利区,并开展野外调查验证。【研究结果】基于Landsat-8、ASTER两种多光谱数据对铁染、羟基类(Mg-OH、Al-OH)、碳酸盐类矿物信息进行了增强与提取;基于GF-5数据识别出了方解石、钠云母、普通白云母、多硅白云母、明矾石、高岭石、地开石、绿帘石8种蚀变矿物。【结论】结合不同数据源的提取与叠加结果,证实了本文提出的矿化蚀变信息提取方法的可行性。根据野外验证情况综合揭示了该地区发育高硫型浅成低温热液蚀变矿物组合,具有斑岩-浅成低温热液矿床的成矿潜力。本文认为高光谱与多光谱数据相结合有助于后续蚀变分带的分析与更精确的成矿预测,从而更好地服务于矿产勘查工程等领域。

高光谱图谱结合策略检测小麦单粒种子活力

小麦是我国主要的粮食作物,在国民经济发展中扮演至关重要的角色。种子是一切农业活动的基础,种子活力是种子最重要的评价指标之一,高活力的种子拥有良好的田间表现及耐储能力,因此准确鉴别小麦种子活力对我国农业生产具有重要意义。传统种子活力检测技术耗时、对操作人员要求高,且会对种子造成不可逆的损伤。以往利用高光谱成像技术检测种子活力,通常是针对种子批检测,且仅仅利用图像数据或光谱数据中的一种,很少将图谱数据结合用于单粒种子活力检测。为了更深入了解种子活力与光谱的内在联系,高光谱成像的小麦单粒种子快速无损检测研究颇具学术价值。以210粒经人工老化处理过的小麦种子(105粒有活力,105粒无活力)为研究对象,采集种子400~1050 nm波段内的高光谱数据,随后进行标准发芽试验,确保高光谱数据与发芽实验结果一一对应,按照4∶2∶1的比例将数据集划分为训练集、测试集和真实数据集。利用竞争自适应重加权(CARS)算法选择特征波段,最终得到了30个特征波段,且所选特征波段对应了引起种子活力变化的蛋白质、淀粉和脂类等种子内部营养物质。为挑选出最优分类模型,对于全波段和特征波段光谱数据,利用训练集和测试集数据基于SVM、KNN、1DCNN和改进的ECA-CNN机器学习算法分别建立了小麦种子活力预测模型。结果表明,使用特征波段数据建立的模型性能均优于使用全波段数据建立的模型,其中使用特征波段数据建立的ECA-CNN模型性能最好,在避免过拟合的情况下,训练集整体准确率为99.17%,测试集准确率为80%。为避免建模过程对比较分类策略造成影响,利用真实数据集对比整体法和像素法两种分类策略。结果表明,像素法相比于整体法拥有更好的检测效果,整体准确率为86.67%,精确率为92.31%,召回率为80%,均优于像素法。该研究可为快速无损检测单粒小麦种子活力提供科学依据。

基于WorldView-2高分影像信息增强及提取在卡而却卡地区遥感调查中的应用

为进一步探讨WorldView-2高分遥感数据在遥感调查中的应用效果,笔者选择青海卡而却卡地区进行图像信息增强及信息提取方法研究。在分析研究区内不同岩性的光谱曲线特征的基础上,针对遥感解译中目视解译效果不好、图像模糊和对比度不够等问题,提出一系列图像增强处理的方法,这些方法显著增强了岩性、构造的识别效果,可以更好地辅助高分遥感解译工作。同时,通过基于纹理和光谱信息的影像分类方法对大理岩、岩浆岩信息提取研究,可以较准确地圈定大理岩、岩浆岩岩性界线,并通过与现有地质矿产资料进行对比,该方法形成的岩性-构造解译图反映的岩性、构造等信息更加详细、丰富。

多光谱信息下的油纸绝缘局部放电深度学习融合诊断方法

为了获得更丰富的放电特征信息,提高局部放电诊断的效率,提出了一种多光谱信息下的油纸绝缘局部放电深度学习融合方法。首先,基于微型光学传感器件和局部放电光谱分布,构建了局部放电多光谱同步检测平台,通过实验获得了四种放电类型的多光谱数据;然后构建卷积神经网络模型,将不同光谱段的局部放电数据作为模型不同通道的输入,利用通道级融合提取多光谱信号中的有效信息,对油纸绝缘局部放电类型进行准确辨识。结果表明:不同放电类型的多光谱信息可以作为模式识别的有效特征;引入多光谱信息后,本文所提方法的识别准确率可以达到98%以上,相比于仅使用脉冲电流信号有着明显的提升;相比于统计特征参数分析法和深度神经网络,提出的方法对多光谱信息融合的效果更好,识别的准确率更高。

基于NMI-SC的糖尿病混合数据特征选择

针对糖尿病预测精度受高维混合数据影响的问题,提出基于NMI-SC的糖尿病特征选择方法,通过邻域互信息(NMI)计算混合属性特征邻域半径内的联合概率密度,构建相似度矩阵,通过糖尿病特征之间的相似性构建无向图,基于谱聚类(SC)将糖尿病特征切分为多个特征相似组,实现非线性特征间的聚类,根据特征分类重要性选出相似组中的代表特征。并将其与原始特征集在支持向量机分类器上的准确率进行比较,该特征选择方法在删除46个冗余特征后,准确率提高了13.07%。实验结果表明,该方法能有效删除冗余特征,得到糖尿病分类性能优异的特征子集。