面向高效检索的多源地理空间数据关联模型

地理空间信息往往包含矢量数据、栅格数据和文本描述信息,这些信息之间通常相互联系.如何快速、全面检索和定位这些相关联的信息,是地理空间信息应用中的新需求.为提高地理空间数据检索和分析的性能,该文提出一种面向高效检索的多源地理空间数据关联模型MSGCM.该模型通过提取多源地理空间数据空间信息、语义描述信息、内容描述信息及其关联关系,构建特征要素图,并基于关联模式将多源地理空间对象融合到统一空间中.通过计算不同对象之间的关联强度,构建类似图的关联模型.为提高模型构建效率,提出了一种基于特征索引的分块构建方法.与已有方式相比,MSGCM模型可以有效支持多源地理空间信息的关联,进而能够支持地理空间信息查询、分析及综合展现等多种地理空间应用.实验及分析表明,MSGCM可以有效提高多源地理空间信息关联检索结果的多样性,并具备一定的可扩展性.     

基于栅格的GIS三维空间数据模型

空间数据模型是描述GIS空间数据组织和进行空间数据库设计的理论基础，它反映了现实世界中空间实体之间的相互关系，是空间数据处理和空间数据挖掘的核心问题。文章论述了空问数据模型的表示形式，重点讨论了基于栅格的三维空间数据模型，指出了栅格矢量空问数据模型集成化的必要性以及集成化表示方法，给出了将矢量数据转换为栅格数据的欧氏距离变换与算法，为基于栅格空间数据的空间数据处理和空间数据挖掘奠定了基础。     

基于网格模板的栅格与矢量叠加统计算法

为了避免矢量栅格数据叠加统计分析中采用逐像元“扫描”判断法所导致的处理速度缓慢的问题，提出将矢量多边形底图切分成具有一定水平、垂直空间尺度的多行、多列网格模板，保存该模板文件，并在模板的基础上进行快速统计叠加的办法。实质是将矢量栅格数据叠加计算间接转换为计算简便的网格格点与栅格数据之间的叠合，从而提高了计算速度。由于底图格点大小及位置与栅格数据不能有效配准而导致“错位”，该文重点给出了解决此类问题在3种不同匹配方式下的处理算法。     

移动点对象与参考地物空间拓扑关系

针对二维欧氏空间内的平面移动点对象，该文基于“9交模型”，定义了参考地物的空间拓扑关系模型，并将该模型表示为3个连续观测时刻的拓扑关系的三元组，给出了移动点对象与参考点、线和面地物的空间拓扑关系的定性表达。该模型可以清楚地刻画移动点对象相对于参考地物的运动变化情况，对于智能交通、移动位置服务等应用具有重要的意义。     

基于尺度的时空RBAC模型

空间技术的广泛应用,使得空间数据的安全日益重要.空间数据库的访问控制已经成为目前国内外研究的热点问题.现有的空间数据访问控制模型主要分为2大类:一类是对RBAC模型扩展,加入空间位置的判断,提供基于位置的服务,但这类模型缺乏对空间数据自身特性的分析;另一类是基于矢量数据或栅格数据提出的自主访问控制模型,这类模型分析了空间数据的特征,如几何表示、拓扑关系等,但是缺乏一个通用的模型.在传统RBAC模型中扩展了空间位置、时间和尺度等基本数据元素,提出基于尺度的时空RBAC模型(STS-RBAC).通过在角色激活中加入位置和时间的约束条件,增加了空间信息使用的安全性;针对矢量数据和栅格数据定义尺度的概念,增强了访问控制策略的表达能力;同时定义操作集、尺度和空间对象间的关系,使得相关授权信息具有更为简单的表达形式和自动推导特性,减少了重复存储和访问次数,提高了权限判定效率.     

融合雷达数据与地理信息系统的归并算法

融合算法将“不具备完整地理信息”的雷达栅格数据转换为能被GIS平台显示、处理及分析的矢量面几何体。算法从左到右、从上到下扫描雷达栅格矩阵,以相邻点的回波等级分布情况作为网格区域归并的判断依据,处理“左右一致”、“左右、上下一致”、“左右、左上、上下一致”、“左右、上下一致,而左上不一致（岛）”等的相邻栅格点情况,最终将大量的规则雷达栅格数据点归并为数量较少的不规则矢量面,从而提高雷达数据的显示、处理速度,并能借助GIS空间算子的操作,实现与空间点、线、面地物的空间统计分析。实验分析结果显示融合结果正确,对雷达观测的原始结果值在空间上不产生形变;算法处理速度较快,能满足实际应用的要求。而针对雷达矢量面数据空间分析及计算的检验结果表明雷达栅格数据在转换为矢量格式后确实能完成在原始数据格式下很难,甚至无法完成的空间统计分析功能。     

基于GPU Raycasting算法的矢量/栅格混合绘制研究*

Raycasting体绘制算法由于成像质量高而被广泛应用于体数据的可视化,但当线、面表达的矢量数据和三维栅格表达的体数据同时绘制到同一场景时,由于绘制方法的差异会造成矢量和栅格数据空间遮挡关系不一致。在GPU实现Raycasting算法的基础上,通过矢量和栅格数据先后绘制,采用FBO离屏绘制等技术将矢量数据绘制到深度缓存纹理并在体绘制采样和融合中统一考虑矢栅颜色融合。实验结果表明,该算法在矢量数据非透明模式下能正确处理矢量栅格数据的混合绘制。     

3DGIS中空间数据可视化的研究与应用

空间数据可视化是当前3DGIS领域的研究热点之一。提出了一种面向对象的空间数据可视化模型，基于金字塔模型来组织和调度DEM（数字高程模型）、DOM（数字正射影像）等栅格数据，并引入矢量数据参数化3D建模思想，实现了矢量地物数据的三维建模与可视化。在此基础上，借助虚拟现实技术，基于OpenGL实现了一个3DGIS原型系统，具备空间数据的真实感可视化表达、3D场景实时漫游、交互查询等功能。最后给出了基于该原型系统为军事与民用领域开发的应用实例。     

基于模糊粗糙集的GIS数据转换方法研究

GIS按照数据结构可以分为矢量数据和栅格数据两种。GIS的空间建模大部分是基于栅格数据进行的,因此,对于一些在矢量结构中表达的属性（如人口密度等）,需要转换成栅格数据。传统的数据转换方法一般都是面积占优法,这种方法没有充分考虑空间数据的模糊性问题。采用模糊粗糙集的理论,对矢量多边形数据转换成栅格数据的方法做了探讨。基于模糊粗糙集的数据转换方法考虑到了多边形数据的边界模糊性,利用隶属函数来确定多边形边界的不确定性。     

支持栅格数据的GSQL扩展研究

空间数据库已经成为实现海量空间数据管理的高效手段，在空间数据库设计和实现中，定义GSQL是其关键。目前，针对于矢量数据访问已经有许多学者进行了研究，但对于地理栅格数据，其访问和操作接口却尚无相关规范，为此参考OpenGIS，对该方面进行了研究，即首先定义了与栅格相关的抽象数据类型，其分别为Pixel(栅格点)、RasterRegion(栅格区域)、RasterCoverage(栅格覆盖)，它们之间存在聚合关系；然后对每种数据类型进行了定义，包括其数据对象以及相关操作；进而，基于上述抽象数据类型，给出了支持栅格数据访问的GSQL—R描述以及相关实例。通过实例表明，GSQL—R能较好支持矢量／栅格一体化的数据访问和操作。     

栅格数据地理信息系统(Raster GIS)关键技术分析

数据是地理信息系统的基础,强大的地理信息分析功能对地图数据有很高的要求.国内外大多数应用型地理信息系统都是基于矢量地图数据格式的.这样的系统在提供便利操作的同时,增加了系统开发成本和开发周期.如果建立基于栅格数据的地理信息系统,不需要数字化编辑处理地图数据,节省了矢量化数据的成本,并且地图显示更为形象、直观.在分析栅格数据和矢量数据的基础上,探讨了基于栅格数据的地理信息系统体系框架,并就基于栅格数据的地理信息系统开发中的基本地图操作、属性查询、地图分析等关键技术进行研究,提出了一些思路和解决办法.同时,将该技术成功应用到军事演习系统中,取得了良好的效果.     

Vector versus raster: a functional comparison of drawingtechnologies

Raster drawing systems have emerged as an alternative to more traditional vector-based CAD drafting systems. With engineering drawing in mind, the author compares the functional capabilities of these technologies and explores means for integrating the best features of both. The characteristics examined are not those of any particular raster or vector implementation but are derived from the underlying technologies. Three means are provided to blend the capabilities of raster and vector systems, each resulting in drawing systems that have capabilities beyond those of either system alone     

一种光栅图像地图转化成矢量数据地图的方法

矢量化光栅地图是一个重要而又远未解决的问题。该文通过对分布式虚拟战场环境中仿真实体及其应用研究的开发应用，对光栅图像地图转化成矢量数据地图的方法进行研究。该方法首先对扫描仪扫描输入的图像进行处理，然后采用一种动态线划的自动、半自动跟踪算法，数字化现有地图上的等高线等地貌信息，该方法是利用跟踪地图地理要素线划中心的模式完成跟踪的，较好地模拟了人工矢量化所采用的策略和方法，获得的矢量数据骨架点准确，失真少、处理速度快，与人的直观视觉相符，能用于各类地图的矢量化。     

Inference of disjoint linear and nonlinear sub-domains of a nonlinear mapping

This paper investigates new ways of inferring nonlinear dependence from measured data. The existence of unique linear and nonlinear sub-spaces which are structural invariants of general nonlinear mappings is established and necessary and sufficient conditions determining these sub-spaces are derived. The importance of these invariants in an identification context is that they provide a tractable framework for minimising the dimensionality of the nonlinear modelling task. Specifically, once the linear/nonlinear sub-spaces are known, by definition the explanatory variables may be transformed to form two disjoint sub-sets spanning, respectively, the linear and nonlinear sub-spaces. The nonlinear modelling task is confined to the latter sub-set, which will typically have a smaller number of elements than the original set of explanatory variables. Constructive algorithms are proposed for inferring the linear and nonlinear sub-spaces from noisy data.     

基于EMD多模态特征融合支持向量机的故障诊断

针对非平稳时间序列信号,提出一种基于经验模态分解(EMD)的特征提取和多模态特征融合支持向量机的故障诊断方法,首先对原始信号进行EMD分解,选择能量最大的几个基本模式分量(IMF)并提取其小波包特征,获得对每个IMF独立的特征子集;然后在每个IMF特征子集中训练SVM弱分类器,并根据各特征子集对应的IMF能量权重进行加权融合,获得故障状态的强分类器,将该方法应用于6135型柴油机振动信号故障诊断中,实验结果表明了其可行性和有效性.     

The raster graphics approach in mapping

Graphics hardware technology is expanding from pen plotters and vector screens to raster. The decade for outstanding advancements in raster output devices, in terms of greater resolution, more colors and greatly reduced costs, is upon us.

Ever since the first graphic output devices were developed, discussion as to whether the raster or vector technique is better has occupied computer graphics analysts. There are problems which are easier to solve in one system, but almost impossible to attack in the other. For example, high-precision engineering drafting consists mainly of short line segments, an operation very cumbersome for a raster structure, but simple with the vector technique. On the other hand, solid shaded areas, colored lines of varying thickness and satellite image processing are impossible with vector systems, but natural for a raster display.

Recent years developments have concentrated on devices which do not draw straight lines, but instead use a matrix of dots to build up the picture. This method is the raster technique.     

Raster to polygon conversion of images

Raster graphics systems frequently store both a vector and a raster representation of the image. The vector representation is convenient for data input and editing, while the raster format is required for refreshing the display. Methods of converting polygons to raster format and raster to polygon format are discussed. A simple scheme which uses overlapping polygons is described.     

Topology in Raster and Vector Representation

Egenhofer's nine-intersection, well-known for vector representations, is defined here on a raster, using a hybrid raster model, and then systematically transformed to yield functions which can be used in a convolution operation applied to a regular raster representation. Applying functions, the hybrid raster elements need not be stored. It becomes thus possible to determine the topological relation of two regions, given in raster representation, with the same reasoning as in vector representations. This contributes to the merging of raster and vector operations. It demonstrates how the same conceptual operations can be used for both representations, thus hiding in one more instance the difference between them.     

Integration of remotely-sensed raster data with a vector-based geographical information system for land-use change detection

Methods of land-use change detection are different for raster and vector data types because of the differences in structures of the two data types. Since large amounts of land-use data (derived from existing maps and aerial photographs) are stored in vector format in a Geographical Information System (GIS), there is a need to develop a change detection algorithm for use with vector-formatted data. Since remotely-sensed images are increasingly being used to derive land-use data, it is necessary to integrate raster data with large volumes of vector data already available in a GIS. This necessitates an efficient and effective data integration technique using which raster data are to be integrated with a vector-based GIS. This paper presents a methodology of data integration of remotely-sensed raster data with vector data. A new approach is developed for land-use change detection for use with vector data in a GIS environment. The approach is based on the mathematical concepts of Sets and Groups, and is successfully implemented for the analysis of historical land-use change from 1931 to 1989 in the River Glen catchment, U.K. Algorithms have been developed for automatic derivation of dynamic statistics of land-use. It is shown that this approach can be efficiently adopted for operational use incorporating products derived from both coarse- and fine-resolution remotely-sensed satellite images once these are integrated with the vector-based GIS.     

Bidirectional raster scanning to increase time for vector writing in avionics see‐through displays

See‐through displays (STD) like head‐up displays (HUD) play major role in providing aircraft data to pilot through a visual interface. The data is displayed in form of symbology in pilot's line of sight using vector scanning technique and is deciphered to get flight and mission information. Contrast ratio required to see against bright sunlight at high altitudes is given only by luminance of cathode ray tube (CRT). Therefore, CRT is still used as display source in avionic HUDs, although other devices are also being explored for these applications. Aircraft's night mode operation requires raster display of external scene information along with vector data display due to low visibility during night. It reduces time available for writing vector data during night mode resulting in reduced symbology. Conventionally used raster scanning methods are unidirectional scanning methods. Bidirectional raster scanning (BRS) is an unconventional approach which scans display screen from both directions. In present work, an algorithm for BRS has been developed in VHDL providing more time for vector writing by increasing vertical retrace period (VRP) of raster scanning. Functional verification of design has been performed by simulations carried out through test‐bench and VRP is increased to 5.8304 ms from conventional 1.2800 ms per field.