Nouvelle perception de la spatialité des objets et de leurs relations. Développement d’une modélisation tridimensionnelle de l’information spatiale

ROLAND BILLEN | 3 Juillet 2002

Summary

Our understanding of the world around us extends beyond traditional two-dimensional representations. Embracing the multidimensional nature of our environment is crucial in accurately modeling geographical phenomena. Over recent years, strides in 3D geographical information systems (GIS) have aimed to address this reality, leveraging advancements in computational power and the availability of three-dimensional data from various sources such as GPS, aerial and satellite photogrammetry, and laser scanning.

Despite these advancements, 3D-GIS applications remain somewhat niche, and commercial developments have yet to fully harness the multidimensional nature of spatial information. The challenge lies not only in technical constraints but also in overcoming deep-rooted conceptual paradigms shaped by centuries of two-dimensional cartographic tradition.

Transitioning from 2D to 3D requires a shift in perspective, prompting geographers to delve deeper into spatial foundations previously simplified in two dimensions. By focusing on urban environments as a primary domain, we can explore discrete geographical phenomena and refine our understanding of spatial relationships and object representations in three dimensions.

While topological relations have long been foundational in 2D GIS, they fall short of capturing the complexity of the 3D spatial reference frame. Existing 3D modeling approaches, whether derived from GIS or computer-aided design (CAD), often lack the comprehensive spatial analysis capabilities necessary for multidimensional GIS. Bridging this gap requires an integrated approach that combines the strengths of both disciplines.

A novel contribution of our research is the development of the dimensional model (DM), which characterizes objects based on convexity criteria. This model introduces dimensional elements that define spatial relations beyond traditional topological, metric, and ordinal categories. Offering greater flexibility and precision, the DM represents an intermediate solution that enriches existing spatial data models.

Through the implementation of a prototype tested with three-dimensional urban data, we demonstrate the efficacy of the DM in uncovering dimensional relations between objects and facilitating query visualization. As the field of geographical information science evolves, embracing this new spatial paradigm holds promise for the advancement of multidimensional GIS applications."

Comments

This doctoral research has led to several research themes that were foundational to the development of GeoScITY. Firstly, it enabled a foothold in qualitative spatial reasoning and more conceptual and ontological approaches. Secondly, it paved the way towards work related to (infra)structures of 3D spatial data. Thirdly, from an applied perspective, it confirmed the interest in urban scales ranging from the city to buildings. Additionally, and underlyingly, this required a follow-up on acquisition techniques enabling the acquisition of 3D data and their automatic processing.

Nouvelle perception de la spatialité des objets et de leurs relations. Développement d’une modélisation tridimensionnelle de l’information spatiale

Summary

Comments

Links