支撑向量机及其遥感影像空间特征提取和分类的应用研究

提出了基于支撑向量机（SVM）的遥感影像空间特征提取的新方法，并以SPOT全色波段影像上城市特征信息的提取为应用实例，并与人工神经网络（ANN）等特征提取方法进行综合比较，认为SVM方法不但能够获得比较高的分类精度，而且在学习速度、自适应能力、特征空间高维不限制、可表达性等方面具有优势。     

SVG网络图像标准支持下的地图显示

介绍了一种全新的网络图像标准SVG（Scalable Vector Grcphics），分析了目前网上地图显示的方法，探讨了运用SVG在网上进行地图显示的可能性，提出了其应用体系和结构。     

科学工作流及其在空间决策支持系统中的应用

分析了传统DSS和GIS在处理复杂空间决策中存在的问题,介绍了工作流的概念和方法 ,最后在理论模型上对基于工作流的空间决策支持系统进行了探讨.     

长江防洪决策支持系统——总控管理与人机界面

以长江防洪决策支持系统为例,讨论了总控系统与人机交互界面软件的开发.介绍了总控管理系统的组成及总控与其它各子系统之间的软件集成技术,并提出了一些技术难点的解决方法;描述了基于地理信息系统(GIS)技术的空间和属性数据处理方法,结合图形用户界面(GUI)技术,简述了系统人机交互界面的设计开发方法;讨论了基于电子地图的图形操作方法.     

历史上的潮汐推算机

海洋科学发展史上,开尔文机械式潮汐推算机的出现,堪称是海洋仪器中机械文明的典范,在潮汐预报方面整整统治了四分之三个世纪。由于我国未曾引进与研究,本文对其原理与应用效果作了分析与介绍。     

Dynamic Response of a Machine Foundation in Combination with Spring Mounting Base and Rubber Pad

The effect of two different combinations of a spring mounting base and a rubber pad sandwiched between the machine base and its concrete footing block was examined experimentally. The machine was subjected to steady state horizontal harmonic excitation with the help of the Lazan mechanical oscillator. Experiments were performed by inserting a spring cushioning system between the machine base and its concrete footing. In addition, two additional series of experiments were carried out by placing a rubber pad either just above or below the spring cushioning system. It was clearly noted that an employment of a spring cushioning system, as an active isolation measure, becomes the most effective when the rubber pad is kept just below the machine base. For different combinations of the spring mounting base and rubber pad, two different modes of resonance were clearly observed and the peak amplitudes of vibrations were found to be associated always with the second mode; the predominant frequencies associated with the peak amplitude of vibrations (second mode) were found to be significantly greater than those without the employment of any isolation system in which case only a single mode of resonance was observed. It is expected that the study will be beneficial for assessing the performance of the machine foundations with the employment of springs and rubber pad.     

地震序列类型的确定与现场预报规则的获取

论述了“震后趋势决策支持系统PTDSS”中的知识学习问题。以某些地震参数与序列类型的关系为例,介绍了如何确定地震序列类型以及系统通过FAM模型进行机器学习的方法。通过学习系统得到了一批非常有用的早期判断序列类型的知识。     

FHD 磁力仪分量线圈装置误差对磁偏角观测的影响分析

根据FHD 磁力仪观测磁偏角的原理, 计算了分量线圈不水平导致的磁偏角观测值误差,发现分量线圈仅在EW 方向的倾斜对磁偏角观测产生误差。文中将理论分析结果与实际观测结果进行了对比分析, 发现由于分量线圈底座水平的长期漂移, 仪器的垂直分量及磁偏角观测误差也会发生长期漂移, 长期漂移变化的周期约2 ~ 4 个月, 对垂直分量其幅度约3.5~ 11.0nT, 而磁偏角的幅度约0.1′~ 1.0′, 因此认为在将该仪器观测值用作绝对观测值进行地震异常分析时, 应注意回避这种长期漂移变化。     

磁化强度矢量反演方程及二维模型正反演研究

导了磁化强度矢量层析成像方程,并与磁化强度标量层析成像方程进行了对比. 使该矢量层析成像方程既适用于三维的也适用于二维,既适用于使用磁场垂直分量资料,也适用于使用磁场总强度资料. 本文采用改进的高斯-赛德尔迭代求解磁化强度矢量层析成像方程,在求解方程中引进了与深度有关的权系数. 并对二维模型开展了正反演研究. 当模型层数为二层,且每层51个柱体时,采用零初始模型就能获得较好的磁化强度垂直分量和水平分量反演结果,除了异常体边部外,磁化方向比较可靠. 当模型层数为五层,且每层51个柱体时,采用零初始模型不能得到较好结果,改用接近背景值的均匀初始模型,反演的磁化强度垂直分量和水平分量大致接近真实,但不能刻划某些细节.     

Landslide susceptibility mapping based on Support Vector Machine: A case study on natural slopes of Hong Kong, China

The Support Vector Machine (SVM) is an increasingly popular learning procedure based on statistical learning theory, and involves a training phase in which the model is trained by a training dataset of associated input and target output values. The trained model is then used to evaluate a separate set of testing data. There are two main ideas underlying the SVM for discriminant-type problems. The first is an optimum linear separating hyperplane that separates the data patterns. The second is the use of kernel functions to convert the original non-linear data patterns into the format that is linearly separable in a high-dimensional feature space. In this paper, an overview of the SVM, both one-class and two-class SVM methods, is first presented followed by its use in landslide susceptibility mapping. A study area was selected from the natural terrain of Hong Kong, and slope angle, slope aspect, elevation, profile curvature of slope, lithology, vegetation cover and topographic wetness index (TWI) were used as environmental parameters which influence the occurrence of landslides. One-class and two-class SVM models were trained and then used to map landslide susceptibility respectively. The resulting susceptibility maps obtained by the methods were compared to that obtained by the logistic regression (LR) method. It is concluded that two-class SVM possesses better prediction efficiency than logistic regression and one-class SVM. However, one-class SVM, which only requires failed cases, has an advantage over the other two methods as only “failed” case information is usually available in landslide susceptibility mapping.