一种面要素语义位置的GeoSOT网格定位方法

为解决传统地图查询方法中对复杂地址对象难以定位的问题,该文面向地理网格的时空大数据分析需求,结合面要素特点与语义分析,提出一种面要素语义位置的GeoSOT网格定位方法。首先通过最小外包矩形(Minimum Bounding Rectangle,MBR)将面要素以九宫格形式分割成9个区块,单独计算各区块的网格编码集合;然后提取位置语句中的名称与空间关系,根据具体情境进行直接查询,或利用网格编码计算目标位置;最后根据语义考虑是否对编码结果做差集运算,从而得到目标的GeoSOT网格定位范围。实验表明,对比现有的地图服务,该方法能够实现语境更加复杂的位置定位,具有可行性和有效性,其提供的未知要素推理定位方法能够有效补充、扩展原有的面要素数据集,可为后续针对面状区域的时空大数据分析提供网格化的数据资源。

A GeoSOT Grid Location Method for Semantic Location of Area Features

In order to solve the problem that it is difficult to locate the objects with complex address in traditional map query methods,this paper proposes a GeoSOT grid location method for the semantic location of area features based on the characteristics of area features and semantic analysis.Firstly,each area feature is divided into nine blocks by using minimum bounding rectangle(MBR)in the form of nine grid,and the GeoSOT grid code set of each block is calculated separately.Then,the name and spatial relationship in location statements are extracted to query the region directly according to the specific situation or to calculate the target position by using GeoSOT grid codes.Finally,the possible regions of targets are marked according to the difference-set of the GeoSOT grid code set.Experiments show that compared with existing map services,this proposed method can locate locations with more complex context,and it is feasible and effective.The method can effectively supplement and expand the data set of original area features,and it provides grid data resources for the subsequent spatial big data analysis which is the foundation of information analysis.