Automatic Building Aggregation Supported by Knowledge Graph

Abstract

ABSTRACT Large‐scale building aggregation presents a significant challenge in cartographic generalization. Current methods have limitations in detecting building proximity relations efficiently and considering spatial structural characteristics. These limitations result in suboptimal aggregation efficiency and the lack of spatial structural characteristics in the final results. To address these shortcomings, we proposed a knowledge graph‐supported building aggregation method. First, we delineated the building spatial structural relations and identified building proximity edges. Second, we constructed a knowledge graph that expressed building proximity relations, spatial structure relations, and other relevant information. Third, considering the cartographic requirements, we designed reasoning rules to efficiently infer the relation between proximity edges and spatial structure. Finally, based on the reasoning results, we employed the proximity edges interaction projection method and the proximity edges angle bisector method to construct the aggregation polygons. To validate the effectiveness of our method, we conducted experiments using Shanghai building data. The results demonstrated the superiority of our approach in terms of both efficiency and quality. Specifically, our method exhibited higher efficiency compared to traditional proximity graph‐based approaches and demonstrated adaptability to large‐scale data. Moreover, it effectively achieved accurate aggregation of buildings under different structural relations and aggregation thresholds.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 3 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.