机构地区: 南京师范大学地理科学学院虚拟地理环境教育部重点实验室
出 处: 《地球信息科学》 2009年第3期292-298,共7页
摘 要: 当前,虚拟地理环境(VGEs)的发展已经走向以数据库和模型库为双核心的知识共享平台阶段。地理数据及模型共享研究是虚拟地理环境发展过程中,不可逾越的难点问题;由于对地学分析及地理建模的研究还处于探索阶段,故适合于VGE下模型集成共享的多源异构数据集成研究还鲜有涉及。本文针对地理数据异构的特殊性,分析了虚拟地理环境中多源数据的异构特性,主要体现在数据含义、模式、数据模型、词法及语义等五个方面。另外,提出了面向虚拟地理环境的数据表达模型(DRM),该模型提供了一种公共的数据组织方法,使用户可以无歧义地定义地理数据,解译其他各种数据,并研究了多源异构数据的集成方法;同时,提出了多源异构数据集成的流程及体系框架,主要包括地理数据解析、地理模型与数据的交互、模型解析、运行及结果输出等;并重点分析了多源异构数据,及其到模型集成的转化过程。最后,开发了原型系统和多源异构数据的集成验证。实验结果表明,该方法能够较好地集成多源异构数据。 Virtual geographic environments (VGEs), which as the opposite of the real environments are the reflection of real environments in mental scene. It is not just a framework which emphasizing data sharing in GIS, and has become a knowledge sharing platform which emphasizing both database and model database simultaneously. As a research platform for earth system science, VGEs are oriented to geographic investigation and analysis of multidomains. It is a very meaningful tool for geographic modelling, simulation and geographic problem solving. The world of today heavily relies on spatial data to manage the natural and built environments, also to construct VGEs. In this paper, the characteristics of mutli-sourced heterogeneous spatial data in VGEs were discussed, and heterogeneity in spatial data can be divided into five types, namely data values, schema, data model, syntactic and semantic heterogeneity. A Data Representation Model oriented to VGEs (DRM) which provides a common data model to define a data representation structure for traditional data and spatial data was designed in this paper. The DRM not only provides a clear description of geographic data, but also defines the relationships between geographic data. The processes and framework of mutli-sourced heterogeneous spatial data integration were proposed based on the above discusses. The integration consists Of the following subparts: geographic data parsing, interaction between geo-data and geographic model, and geographic model parsing, running and results output. Also, spatial data parsing and the communion between mutli-sourced heterogeneous spatial data and geographic model were discussed especially. Lastly, the platform system of multi-sourced heterogeneous spatial data integration was designed and developed under the VS. NET2005 development environment combined with XML technology, and it can solve the bottleneck problems arise in geographic information resources which under heterogeneous environment. Some experiments were designed and discuss