大尺度自然地理实体边界智能化提取方法OA北大核心CSTPCD
Intelligent Boundary Extraction Method for Large-scale Physical Geographical Objects: Taking Dabie Mountains as an Example
为了解决语义明确但空间位置与分布范围并不明确的自然地理实体(PGO)在地图中的分布范围或边界表达问题,提出一种PGO边界的智能化提取方法.首先利用给定的语义词,自动搜索网络地图大数据;接着在顾及PGO的连续空间分布特性基础上,运用空间聚类算法确定PGO的大致范围;然后利用PGO的几何特征(如山地的起伏变化),运用特征识别算法,进一步确定自然实体的分布范围和边界.考虑到此类实体的复杂性,仅以山地(大别山)为例进行实证研究,验证所提方法的有效性.
In order to solve the problem of expressing the distribution range or boundary of physical geographical objects (PGO) in maps with determinate semantics but indeterminate spatial location or distribution range,an intelligent extraction method for PGO's boundary is proposed.Firstly,the given semantic words is used to search big data of the network map.Secondly,the spatial clustering algorithm is used to determine the approximate range of PGO.Finally,considering the geometric characteristics of PGO,such as the undulations of mountains,a feature recognition algorithm is used to further determine the distribution range and boundaries of PGO.Taking into account the complexity of such objects,only the mountain (Dabie Mountains) was taken as an example to proved the effectiveness of the proposed method.
杨涵珺;孙敏;楼夏寅;杨仕浩
北京大学遥感与地理信息系统研究所,北京 100871
大尺度自然地理实体地图大数据智能化提取
large-scale physical geographical objectsmap big dataintelligent extraction
《北京大学学报(自然科学版)》 2024 (004)
673-681 / 9
内蒙古自治区科技重大专项(201701B)资助
评论