Premium
An automated method for the selection of complex railway lines that accounts for multiple feature constraints
Author(s) -
Li Chengming,
Liu Xiaoli,
Wu Wei,
Wu Pengda
Publication year - 2019
Publication title -
transactions in gis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.721
H-Index - 63
eISSN - 1467-9671
pISSN - 1361-1682
DOI - 10.1111/tgis.12575
Subject(s) - marshalling , generalization , railway line , line (geometry) , robustness (evolution) , computer science , selection (genetic algorithm) , cartographic generalization , process (computing) , data mining , feature (linguistics) , topology (electrical circuits) , extensibility , algorithm , artificial intelligence , engineering , mathematics , geometry , civil engineering , mathematical analysis , biochemistry , chemistry , linguistics , philosophy , electrical engineering , gene , programming language , operating system
The automatic selection of railway marshaling lines is an important procedure in the generalization of road network data. The line selection results must be geometrically consistent with the topological features of the external/internal structural and topological connectivity of the railway stations. When current methods are used to process the lines at railway marshaling stations, which exhibit crisscrossing, dense, non‐hierarchical, and complex characteristics, the selected lines tend to be topologically and structurally erroneous. In this article, we propose an automated method for the selection and generalization of complex railway lines that accounts for multiple feature constraints. First, we identify four types of line structures in accordance with the spatial and topological features of railway stations. Then, based on length and spacing thresholds specified for the generalization process, line selection is performed while accounting for constraints regarding structural robustness and topological extensibility. Finally, we report the validation of our method on a dataset relating to Chengdu Railway Station, China. The results indicate that our method successfully preserves the external and internal structural features and connectivity of railway stations.