Research On Determination Of Object Layout Region In Visualization Of Spatial Relation

Visualization of Spatial Relationship is the core of Text to Scene. Realization on the qualifying of the object spatial layout region is a very important step of Visualization of Spatial Relation in Text, whose aim is to make spatial relation quantitative. This thesis is a very important middle part of the Nature Science Foundation"the 3D visualization of spatial relationships in text based on ontology", which includes three parts: the extraction of spatial relation, the quantification of spatial relation, and object-layout of the spatial relation.The task of this paper is to make spatial relations quantificational, but, we need to learn objects' property and geometry knowledge in order to transform from qualitative to quantitative. So people's spatial common sense is made as start point to construct a few knowledge base,, and then the computing models of spatial relation is constructed based on the knowledge base. At last, the layout area is transferred to optimization layout model so as to get layout sense by computing models.The content of this paper is divided into the following two aspects:1. Knowledge base construction, considering task complexion, we put forward recovering the orient word in the spatial relation; constructing knowledge base, including orientation knowledge base, entity knowledge base, graphic knowledge base, constraint knowledge base. Orientation knowledge base contains the basic orient class information, which can be used to construct basic computing models. Entity knowledge base provides computing models information, such as, life or not, fly or not and so on. Besides, there are some geometry properties such as normal line, orientation and so on.2. Computing models construction, based on the content of knowledge base, different computing model is constructed according to different basic spatial relation. The computing models are composed of basic orientation models and constraint spatial relation models, and the latter is based on the constraint knowledge base, which is composed by spatial common sense. After area computing, we can obtain rational area by spatial reference, then transfers to optimization layout model.