欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
提出了一种基于对自组织映射聚类的道路网网格模式识别方法。以道路网中的网眼为基本单元，从网眼自身形状特征、相邻网眼的形状特征以及与周围网眼的关系等方面定义了5个参量。将由5个参量描述的网眼及由CRITIC方法导出的参量权重作为自组织映射的输入，经过训练，运用K—means方法对神经元码书向量进行聚类。对深圳市道路网数据进行了实验和对比分析，结果表明该方法能有效识别网格模式。
英文摘要：
Pattern recognition in road networks plays an important role in map generalization, data matching and spatial analysis. A grid is characterized by a set of mostly parallel lines, crossed by a second set of parallel lines at roughly right angles. A grid pattern is one of the most typical patterns in road networks. This paper proposes a novel approach for the recog- nition of grid patterns based on the clustering of self-organizing maps （SOM）. Five parame- ters are defined to characterize a mesh from the shape and the context perspectives. The me shes constitute vectors in attribute space. These vectors then are used to train a SOM. The neurons of the SOM correspond to a set of meshes with similar properties. The K means al- gorithm is used to cluster the codebooks after a training process. An experiment and compar- ative analysis for the Shenzhen road network was conducted. The results validate that the proposed approach effectively recognizes the grid pattern and the limitations of the proposed method are discussed.
同期刊论文项目
同项目期刊论文
Methods and strategy for modeling daily global solar radiation with measured meteorological data - A
Impacts of transportation routes on landscape diversity: a comparison of different route types and t
Quantifying determinants of cash crop expansion and their relative effects using logistic regression
Spatially non-stationary response of ecosystem service value changestourbanization in Shanghai, Chin
Land use changes to cash crop plantations: crop types, multilevel determinants and policy implicatio
Enhanced adsorption of aqueous perchlorate on quaternary ammonium chloride surfactant-modified activ
期刊信息
位置：