【作者】 杨斌；

【导师】 何政伟；

【作者基本信息】 成都理工大学 ， 地球探测与信息技术， 2009， 博士

【摘要】 “数字山地”提出源于“数字地球”大背景下孕育而生的,它的建立旨在促进山地学和可持续山地发展,为协调山地区域开发和环境保护,提供数据、信息和方法支持,从而为山地资源的开发、山地环境的保护、山地灾害的防治、山地安全和发展提供决策支持,为山区的资源开发利用、山地灾害监测评价、山地生态环境保护和山区社会经济建设等方面发挥重要的作用。由于山地是个多要素、多层次、多资源和多领域的复杂巨系统,对山地的认知和了解有助于建设科学合理的“数字山地”平台。源于此,本文提出了山地本体的科学概念,旨在用形式化语言形式去表达山地的概念、属性、内涵及空间数据组织方式,并建立“数字山地”中山地概念化本体模型,还对山地本体的逻辑结构和构建方法进行了探讨,提出了三层本体结构下空间数据集成方案。最后在山地本体框架体系下,对山地岩性特征进行了分类,并基于多因子数据对山地数字化分类方案进行了深入探讨,论文的主要研究内容和取得的主要成果如下:(1)运用本体理论思想对山地的认知层次进行了剖析,提出了山地本体概念的表达形式,分析了空间本体、领域本体与山地本体的联系与区别。并从“数字山地”建设角度出来,探讨了“数字山地”中山地本体的概念分类体系、语义场及其特征,从面向本体的山地空间数据库构建形式角度探讨数据库的构建过程。(2)在详细分析本体及地理本体的逻辑语言和结构的基础上,从空间数据与描述逻辑两个内在因素方面对山地本体的逻辑结构进行了剖析,提出了宏观和微观两个层次的山地本体概念化逻辑结构原型。并提出了基于本体论的山地三维空间关系分析方法,最后基于本体建模理论构建了山地本体概念化模型。(3)详细总结了地理空间数据集成的概念、类型及其存在问题,分析比较了现有三种基于本体结构下的不同集成构架方案。在此基础上,提出了基于混合架构下的三层本体集成方案。利用此方案,设计了一种面向“数字山地”的空间数据集成框架体系,并对空间地理信息数据进行了分类方法研究。(4)利用Hutchinson算法将等高线地形数据生成1:25万四川省DEM数据,分别提取出绝对高程、相对高程、坡度、坡向、地面粗糙度、地形起伏度以及沟壑密度7个因子数据,并对这7类因子进行了分类方法研究。在针对于沟壑密度的提取过程中,总结建立了一套提取沟壑密度的空间数据模型,并选取四川省122个县作为试验样区,提取出四川省沟壑密度因子图层。(5)针对GIS分类数据中存在的空间数据投影信息多样性问题,总结了一套空间数据投影转换的新方法,并以北京54坐标向西安80坐标转换为例进行了详细的理论分析与实践说明。该方法先根据不同坐标系下4个相应坐标点的对应坐标值,在用Delphi语言开发的七参数转换软件中求取坐标转换的七参数值,并在ArcGIS平台下实现坐标系统的转换。(6)建立一套由相对高程、绝对高程、坡度和地面粗糙度四因子综合分析的山地与非山地分类体系,在此基础上提取出四川省山地面积,对山地与非山地范围的界定提出了一种新的划分方法。同时采用山地综合评价公式,对山地地貌等级特征进行特征分类。并选取了相对高程、坡度、地面粗糙度、沟壑密度和地面起伏度5个因子,利用层次分析(AHP)和多因子加权法,提取出7类不同的山地地势起伏类型。采用两两组合叠加分析法,对地貌类型数据和地势起伏类型数据进行叠加分析,提取出50种不同类型的山地综合类型数据,实现了山地数字化分类新方法研究。更多还原

【Abstract】 "Digital Mountains", which was made from the "Digital Earth", aims to promote the sustainable mountains development in the mountains, to provide data, information and methods of support for the coordination of regional development and environmental protection. It plays an important role in the development of Mountains resources, the protection of Mountains environments, Mountains disaster prevention, security and development of Mountains areas in decision-making support for the development and utilization of Mountains resources, monitoring and evaluation Mountains hazards, Mountains eco-environmental protection and socio-economic construction of the Mountains. The Mountains is a multi-element, multi-level, multi-resources and multi-field system, it is complex and giant. The cognition and understanding of the Mountains is helpful for the construction of the scientific and reasonable "Digital Mountains" platform. From this, this paper presents the scientific concept of the Mountains ontology, to express the concept, properties, and spatial data organization with the form of formal language, and establish the conceptualization of ontology model in "Digital Mountains". Moreover, this paper also discuss the logic structure and building methods of the Mountains ontology thoroughly , put forward a integrated spatial data classification program in three level ontology structure. Finally, in the Mountains ontology framework, this paper classifies the lithologic characteristics of the Mountains systematically. In addition, we do profound study of digital classification schemes of Mountains which is based on multi-factor data. The main content and fruit of this paper is as follows:(1) With the concept of Mountains characteristics and attributes, this paper analyze the field of geographic features and characteristics of Mountains with profound research and analysis, then analyze the cognitive level of the Mountains with ontological theory of thinking, and create the expression of the ontological concept of Mountains. The connection and difference of space ontology, domain ontology and Mountains ontology was also analyzed. From the construction of "Digital Mountains", we introduce the classification system and the semantic field’s characteristics of geographical ontology in "Digital Mountains". Furthermore, we analyze the spatial entity and the concept of expression of Mountains. Finally, we design the construction form of Mountains spatial database.(2 ) This paper analyzes the logical structure of Mountains ontology, and put forward the construction methods of Mountains ontology. It analyzes the logical structure of Mountains ontology with two aspects of internal factors—spatial data and logical description, which is based on the detailed analysis of the logical language and structure of ontology and geographical ontology. So we put forward conceptual logical structure prototypes of Mountains ontology in macro-level and micro-level. This paper sums up the research of the existing construction methods of geographical ontology, and put forward the method of three-dimensional spatial analysis of Mountains, which is based on ontology. Finally, we also build conceptual model of Mountains ontology with the ontology modeling theory.(3 ) This paper constructs three level integrated program of ontology with hybrid structure. The theoretical study realizes the integration of spatial data classification of "Digital Mountains". To solve the problem of the semantic integration of spatial data heterogeneity in GIS, this paper sums up the concept, types and problems of Geo-spatial data integration by means of the classified applications of Geo-spatial data. It analyzes and compares some different integrated framework which is based on the existing ontology structures. On this basis, we put forward a three-level integrated program in the hybrid structure. Meanwhile, we design the spatial data integrated framework of "Digital Mountains" by the program, and study the classified methods of spatial and geographical information data. We discuss the realized process of classification of the Mountains lithologic factors by the means of the macro-concept prototype of the Mountains ontology.(4 ) By means of Hutchinson algorithm, this paper makes 1:250000 DEM in Sichuan Province from the topographic contour data. To realize the specific applied target, we carefully analyze and descript some factors, such as absolute elevation, relative elevation, slope, aspect, topographic characteristics and hydrology. They are made from the DEM by relative means. Based on the detailed summary of various conception of Mountains, this paper builds a classified system of Mountains and non-Mountains, which is worked comprehensively by three factors of relative elevation, absolute elevation and slope. Then it gets the distinction of Mountains and non-Mountains in Sichuan Province. In the extraction process of gully density, this paper builds a data model which could get gully density automatically. In Sichuan Province, 122 counties are regard as test plots so that the factor layer of gully density in Sichuan Province is yielded.(5 ) There is a variety of projection information of spatial data in GIS classified data. To solve the problem, this paper sums up a new method of spatial projection. To take that the Beijing 54 coordinate transform to the Xi’an 80 coordinate as an example, we analyze and practice with detailed theory. The method resorts to the parameter conversion tools which are developed by Delphi language. It gets seven parameters according to the corresponding coordinates of four points in ArcGIS desktop.(6) I t builds the classification system by the four classified factors of elevation, relative elevation, slope, surface roughness,It provide a new method of the Mountains and un- Mountains. It classifies the Mountains topography grading systematically by means of the formula of Mountains comprehensive evaluation. The five classified factors of relative elevation, slope, surface roughness, gully density and surface relief are chosen to overlay by means of AHP and multi-factor rating weighted method so that 7 different types of data of Mountains terrain relief was gain. The research classifies geomorphology type and relief by means of the overlay analysis of two factors; it gets 50 kinds of comprehensive data of Mountains. It achieves a new method of Mountains digital classification.更多还原