Automatic identification and feature recognition of the metro-led underground space in China based on point of interest data

Abstract

Metro-led underground space (MUS) plays a crucial role in modern underground space utilisation. Recent studies have shown its great potential for high-quality urban development. However, limited evidence about MUS was available on a national scale, resulting in incomplete and unsystematic knowledge of MUS utilisation. The interaction relationship between MUS and the surrounding built environment also remains unclear. To fill the research gap, an automatic method for MUS identification and development features extraction was proposed based on point of interest data. We applied the method to identify the MUS in 28 Chinese cities and estimated the development status of MUS in China for the first time. The nationwide statistics of MUS and correlation analysis of development features were conducted. Results show that complex MUS (CMUS) share is significantly lower than that of simple MUS. Besides, CMUS development in China is primarily dominated by public transport and does not have a solid functional link to its surroundings. The comparative analysis of MUS development in four primary urban agglomerations was also conducted, and their development characteristics were discussed. The study aims to expand the planning toolkit and construct the MUS database, which sheds light on the data-driven planning for MUS.

Keywords

• Development features
• Metro-led underground space
• Point of interest