Characterization and Detection of Building Patterns in Cartographic Data: Two Algorithms

Building patterns are important features in applications like automated generalization and spatial data mining. Many previous work has however focused on a few specific patterns (i.e. collinear pattern), while many others are less discussed. This paper proposes a comprehensive typology of available building patterns through the study of existing maps, and discusses their characteristics. This typology includes collinear, curvilinear, align-along-road, grid-like and unstructured patterns. Two algorithms are presented to detect align-along-road and unstructured building patterns, which are tested against a topographic dataset of the Netherlands.