Intégration des algorithmes de généralisation et des patrons géométriques pour la création des objets auto-généralisants (SGO) afin d'améliorer la généralisation cartographique à la volée

Authors: Sabo, Mamane Nouri
Advisor: Bédard, YvanMoulin, Bernard
Abstract: With the technological development of these past years, geospatial data became increasingly accessible to general public. New applications such as Webmapping or SOLAP which allow visualising the data also appeared. However, the dynamic and interactive nature of these new applications requires that all operations, including generalization processes, must be carried on-the–fly. Automatic generalization has been an important research topic for more than thirty years. In spite of recent advances, it clearly appears that actual generalization methods can not reach alone the degree of automation and the response time needed by these new applications. To improve the process of on-the-fly map generalization, this thesis proposes an approach based on a new concept called SGO (Self-generalizing object). The SGO allows to encapsulate geometric patterns (generic geometric forms common to several map features), generalization algorithms and the spatial integrity constraints in the same object. This approach allows us to include additional human expertise in an efficient way at the level of individual cartographic features, which then leads to database enrichment that better supports automatic generalization. Thus, during a database enrichment process, a SGO is created and associated with a cartographic feature, or a group of features. Then, each created SGO is transformed into a software agent (SGO agent) in order to give them autonomy. SGO agents are equipped with behaviours which enable them to coordinate the generalization process. As a proof of concept, two prototypes based on Open Source technologies were developed in this thesis. The first prototype allows the creation of the SGO. The second prototype based on multi-agents technology, uses the created SGO in order to generate data on arbitrary scales thanks to an on-the-fly map generalization process. Real data of Quebec City at scale 1: 1000 were used in order to test the developed prototypes.
Document Type: Thèse de doctorat
Issue Date: 2007
Open Access Date: 13 April 2018
Grantor: Université Laval
Collection:Thèses et mémoires

Files in this item:
24670.pdf3.22 MBAdobe PDFView/Open
All documents in CorpusUL are protected by Copyright Act of Canada.