| 摘要: |
| 在空间信息处理中,一些常识空间信息通常结合多方面空间关系,而且这些空间关系是动态变化的.为了有效地表示这些复杂的空间关系,并对其进行推理,提出了一种结合拓扑、方向和大小关系的空间信息处理模型TDSC (topology-direction-size calculus),并基于TDSC 模型提出了处理动态空间关系变化的表示推理框架.首先,利用同对象多属性的方法建立了融合大小、拓扑和方向关系的完备互斥基本关系表示;然后提出了复合表生成算法和推理算法,使得原有模型的表示和推理结果可以直接在新模型中使用.同时提出处理动态空间关系的邻域划分图,给出了邻域划分图的自动生成算法,以及TDSC 模型的邻域划分图.最后给出基于TDSC 模型邻域划分图的表示和推理框架,并结合实例说明框架的正确性和有效性. |
| 关键词: 定性空间推理 常识空间信息 邻域划分图 |
| DOI：10.3724/SP.J.1001.2012.04136 |
| 分类号: |
| 基金项目:国家自然科学基金(60973089, 60873148, 61170092, 41001302, 61103091); 国家教育部博士点专项基金(20100061110031); 吉林省科技发展计划基金(20101501, 20100185, 20090108, 20080107, 201101039); 浙江师范大学计算机软件与理论省级重中之重学科开放基金; 吉林大学“985 工程”研究生创新基金(20080242); 欧盟合作项目(155776-EM-1-2009-1-ITERAMUNDUS-ECW- |
|
| Qualitative Spatial Model Used f or Processing Spatial Information in Commonsense |
|
SONG Xiao-Hua1,2,3, OUYANG Dan-Tong1,2
|
|
1.College of Computer Science and Technology, Jilin University, Changchun 130012, China;2.Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun 130012, China;3.Center of Mathematics for Applications, Aviation University of Air Force, Changchun 130022, China
|
| Abstract: |
| In spatial information processing, spatial information is usually combined with various spatial relationships, which are often dynamic. To represent and reason with these complex spatial relationships effectively, a novel model topology-direction-size calculus (TDSC) is proposed, which is integrated with multi-aspects qualitative spatial relations. Then, a framework for dealing with dynamic qualitative spatial relations is proposed. First, a base relation set which integrated with multi-aspects is constructed. Next, the algorithm of constructing composition table is proposed for reasoning, which allows the reasoning result of original model to still be used in new model. To handle the dynamic spatial relations, the neighborhood partition graph (NPG) is proposed, and an algorithm is give to generate the GNP. Using this algorithm, one can get the new model’s GNP easily. Finally, the framework for handing dynamic spatial relations is proposed, which is based on the new model TDSC and its GNP. An example is used to show the framework is correct and effective. |
| Key words: qualitative spatial reasoning spatial information in commonsense neighbor partition graph |
