Recognizing Building Group Patterns in Topographic Maps by Integrating Building Functional and Geometric Information

Recognizing building group patterns is fundamental to numerous fields, such as urban landscape evaluation, social analysis, and map generalization. Despite the increasing number of algorithms available for pattern recognition, there still a lack satisfactory grouping results due insufficient information only geometric features being provided recognition methods. This study aims provide novel method that combines function information. We specifically focus on process recognizing groups in topographic maps prerequisite subsequent First, functions are inferred using dynamic time warping (DTW) algorithm based Tencent user density data POIs (points interest). Then, two types constrained Delaunay triangulations (CDTs) created each block, from which several spatial indices, continuity index (SCI), direction, distance every adjacent buildings, derived. Finally, block modeled graph grounds derived matrices information, segmentation approach proposed extract groups. A case conducted test approach. The experimental indicate can produce results, given correctness value above 81.63% our area. Comparative studies reveal without an ineffective when buildings with different close other. In addition, generalization more line those daily use, they users accurate divisions buildings.