Point-Arc Topology Construction Method Guided By Spatial Index

Topological relationship between spatial entities, which is one of the most basic spatial relationships, is the foundation of spatial reasoning, inquiry, and analysis. Most of the current topological relationship is calculated from the geometric characteristics of spatial objects, the calculation of the topological relations is considered from the perspective of geometric characteristics, without combining similar geography and heterogeneity during the calculation of the topological relations, this makes the calculation method of topological relationships exist room for improvement. The Geographic Information System spatial data model at this stage determines the current topologic calculation process is a "static" pattern, cannot meet the dynamically changing needs; for instance, Geodatabase, whose topological relationship is built on demand, cannot only calculate the changed data. Therefore, based on the analysis of the existing topology calculation process, this paper aims to meet the demand of dynamically updated data, proposes the point-arc topology construction method guided by spatial index. This method hands to build the topology relationship from the perspective of spatial characteristics to improve building efficiencies of static data, on the other hand, achieves dynamic update topological relations.Research results are as follows:(1) Spatial index based on spatial distribution and particle swarm optimizationThe crux of the spatial index is space division. Taking into account the spatial distribution characteristics during the division process is an ideal spatial index. Particle swarm optimization (PSO) has been applied in the clustering problem. This paper uses its clustering characteristics, combines PSO and spatial distribution features to raise the cluster particle swarm optimization (CPSO-TREE). This index not only weighs the amount of data and spatial overlap of minimal geometric parameters, but also takes the spatial data distribution characteristics into account.(2) Point-arc topology calculation based on spatial indexFrom two aspects, topology description and calculation of topological relations, this article researches the present topological relationship calculation methods to analysis the insufficiencies during the construction process, then find the possible optimization direction. That is to construct the topological relationships based on the spatial distribution characteristics. Based on CPSO-TREE, this article proposes a point-arc topology construction method guided by CPSO-TREE. This method builds efficiently for static data; and it only calculates the changed data instead of recalculating the all over data in the previous methods, realizes the topological relations update demand.(3) Prototype SystemAccording to point arc topology calculation method guided by CPSO-TREE, this paper builds a prototype system to calculate topology relationships, experimental results show that the calculating results of this algorithm are correct, the computational efficiency is better than the renowned GIS software such as ArcGIS, SuperMap and GEOS.