【作者】 王振宇；

【导师】 刘艳芳；

【作者基本信息】 武汉大学 ， 地图学与地理信息系统， 2005， 硕士

【摘要】 目前基于知识的地图综合已有一定程度的研究,但大都集中在地形数据或地形图综合知识的研究上,针对某一专题的制图综合知识规则更不多见,主要原因在于专题地图类型的多样性以及对专业知识的特殊要求限制了这方面的研究。因此有必要针对某一领域的专题地图分别建立制图综合知识规则,促进专题地图综合知识的系统化、规范化,促进专题地图自动综合的发展。目前,农用地分等省级汇总工作正在全国范围逐渐展开,本文在这种大环境下总结和归纳了农用地分等汇总制图综合中需要运用和遵循的知识,目的旨在建立农用地分等汇总专题图制图综合知识规则。 本文首先概略的论述农用地分等的概念、对象和农用地分等数据的特点;制图综合的基本理论和方法、农用地分等制图综合的特点以及知识的分类、获取与表达。其次在现有的各种技术规范和综合规则的基础上,较全面的总结了农用地分等汇总制图综合过程中需要遵循和运用的主要综合知识,包括基于应用的知识、专家知识、图形和拓扑关系知识、语义知识、过程性知识,并用人工智能中的产生式规则进行了知识表达和知识推理。然后选取部分湖北省农用地分等中的县级分等数据为试验数据,采用人机交互和批量处理结合的方式进行了试验分析,一方面用于验证综合知识体系,同时也为农用地分等自动综合提供了一个实现方案。最后对论文所做的工作以及论文存在的不足之处进行了总结。更多还原

【Abstract】 At present, the knowledge-based generalization have been researched to some extent, but most of them focus on the research of generalization knowledge of topographic data or map, aim at a certain special purpose is rare. The variety of special purpose map and the special needs of expert knowledge restrict the research on this aspect. So it is necessary to establish knowledge rules for generalization of a certain special purpose map one by one, which can accelerate systematization and standardization of special purpose map and development of automatic generalization. At the present time the pro farmland classification integration are spread gradually on a national scale, Under the circumstances, this paper try to establish knowledge rules for generalization of the farmland classification integration.This paper synoptically discusses the conception of farmland classification, characteristic of data, basic theory and methods of generalization and knowledge classification, acquisition and expression firstly. Secondly based on existing practical standard and generalization rules detailed discusses the generalization knowledge for the farmland classification integration, the generalization knowledge is composed of expert knowledge, graphic and topological relationship knowledge, operation knowledge, thematic knowledge etc, and then represent these knowledge with Production-Rule method of artificial intelligence. And then base on the generalization knowledge do experiment with some data of county farmland classification, through the experiment try to prove the knowledge system, at the same time provides an applicable method for automatic generalization of the farmland classification integration.更多还原