A simplification of urban buildings to preserve geometric properties using superpixel segmentation

Abstract

Building simplification is an important research area in automated map generalization. In the last several decades, various methods for building simplification have been proposed by scholars, most of which have concentrated on vector data. However, with the continuous development of computer vision and artificial intelligence technology, some advanced technologies, such as unstructured image analysis and processing, have provided new opportunities and challenges for map generalization. Therefore, in this paper, we propose a new algorithm called superpixel building simplification (SUBS), for simplifying buildings based on image data. In this method, the buildings are first divided into two types by corner detection: buildings with orthogonal features and buildings with non-orthogonal features. Then, the buildings are globally simplified using a superpixel segmentation algorithm for superpixel extraction via energy-driven sampling. Finally, the buildings are locally simplified to preserve their geometric features. For the purpose of evaluation, we used a total of 285 buildings at scales of 1:5000 and 1:10,000 to perform the simplification. Compared with traditional algorithms, the results indicate that the proposed method can produce satisfactory results for the simplification of buildings with both orthogonal and non-orthogonal features and effectively preserve the area and centre of mass of the buildings. In addition, the SUBS method can generate different representation styles of buildings while effectively avoiding self-intersection.