【作者】 顾星晔；

【导师】 过仲阳；

【作者基本信息】 华东师范大学 ， 自然地理学， 2007， 硕士

【摘要】 随着空间信息获取手段和方法的不断多样化，空间数据的更新速度也在不断加快，这使得空间数据走向实时动态化成为必然。目前，地理信息系统所表达的信息只能反映地理要素在某一时刻的空间状态，对于时间序列方面的分析还处于起步阶段。这就使得时态GIS作为GIS研究和应用的一个新领域，正越来越受到人们的关注。时态GIS所主要关注的内容可以大致归结为两点：十是时空数据模型以及时空数据库的建立；二是时空分析技术以及时空数据可视化问题的研究。本文对这两个问题都有所关注，但重点放在时空分析方面。时态GIS的最终目标是在个合理、完善的时空数据库系统的基础上，融合多尺度时空数据，发展智能化数据处理和分析方法，建立层次结构的智能化认知模型，并在具体应用中起到辅助分析、决策的功能。本文在现有相关研究的基础上，对各种时空数据模型、时空分析技术以及时空数据可视化技术及其特点作出了详细分析，并在此基础上针对矢量和栅格两种基础数据格式分别提出了时空变化的算法和模型，并且应用于河口沙洲演变这一具体问题中。本文的主要研究内容和创新之处如下：1．全面分析了基于状态和基于时间的时空数据模型，比较它们各自的特点，并将这两种模型的应用范围与时空分析方法相结合。2．提出了基于矢量数据的时空变化模型，实现了矢量数据在时间轴上的内插功能，并且在原有时间戳不够密集的情况下，能够达到矢量数据随时间平滑、连续演变的视觉效果。3．提出了基于元胞自动机的栅格数据时空变化模型，通过一定的经验规则，实现对地理变化过程的模拟和预测。该模型能有效克服经典元胞自动机的时空局部性和转换规则同质性等问题，符合地理过程的特点。4．将上述模型应用于河口沙洲变化的实例中，取得了较好的效果。更多还原

【Abstract】 Along with the development of spatial information gathering methods, the updating speed of spatial data is much more faster and the relevant analysis is required to be dynamic and real-time. Current GIS systems, which could only represent geography elements at certain time are lack of the ability of temporal analysis. Therefore, temporal GIS (T-GIS) has been a important and focal aspect of GIS research.The contents of T-GIS could be included into two points. The first is the problem of spatio-temporal data model and spatio-temporal database. The other is the problem of spatial-time analysis and visualization. Both of the two problems are concerned in the research work but our most important research is on the latter problem. The main object of T-GIS is to build intellectualized data processing method and cognitive model which can be used in assisted decision-making based on spatio-temporal database.The research analyzed different spatio-temporal data model and spatial-time analysis techniques. Then spatial-time transformation models based on vector data and raster data are introduced individually. The models have been applied on the problem of shoal changes. The main contents of the research are listed thereinafter:1. Event-based spatio-temporal data model and State-based spatio-temporal data model are analyzed and compared in details. Different spatio-temporal data model are applied in spatial-time analysis on the basis of their own characteristics.2. The spatial-time transformation model based on vector data is introduced and achieves the interpolation ability. The model brings smooth and continuous visual effects in the case of few time-stamps.3. The spatial-time transformation model based on raster data is developed by improving classical cellular automata. The model could simulate geography process by confirming some transformation rules. It also overcomes some disadvantages of classical CA.4. The models above have been applied on the problem of shoal transformation and achieved considerable effects.更多还原