Similar to the intensive developments in Geographic Information Systems (GIS) and spatial databases, the field of urban geosimulation was increasingly used in recent years, with an emphasis on high-resolution applications. This recent tendency does not agree however with the traditionally homogeneous spatial representation provided by regular cellular automata which are commonly used as the spatial structures of geosimulation applications. Several issues are raised in the literature, including reports of limitations in spatial reasoning and a lack of realism related to such spatial representations. The few solutions that, so far, deviate from the ‘dogma of regularity of the spatial representation’ are also subject to recent criticism. In this context, we propose an alternative multi-scale model based on the Voronoi diagram of polygons, which matches geographic features of the urban environment. The proposed theoretical model was developed and tested in the context of a urban geosimulation using high resolution spatial data, in the form of registered parcels of irregular shapes and sizes. The first results of the approach demonstrate the potential of our model in the context of urban geosimulation. Finally, based on our extensive analysis of the main approaches of space segmentation, we propose a schematic method to choose a spatial decomposition suited to urban geosimulation.
| Identifer | oai:union.ndltd.org:LAVAL/oai:corpus.ulaval.ca:20.500.11794/21332 |
| Date | 16 April 2018 |
| Creators | Hins-Mallet, Karine |
| Contributors | Mostafavi, Mir Abolfazl, Moulin, Bernard |
| Source Sets | Université Laval |
| Language | French |
| Detected Language | English |
| Type | mémoire de maîtrise, COAR1_1::Texte::Thèse::Mémoire de maîtrise |
| Format | 193 p., application/pdf |
| Rights | http://purl.org/coar/access_right/c_abf2 |
Page generated in 0.002 seconds