计算机科学与探索 ›› 2024, Vol. 18 ›› Issue (7): 1661-1682.DOI: 10.3778/j.issn.1673-9418.2311083

• 前沿·综述 • 上一篇    下一篇

全景影像在城市研究中的应用进展综述

侯鑫,王艳,王绚,范伟   

  1. 天津大学 建筑学院,天津 300072
  • 出版日期:2024-07-01 发布日期:2024-06-28

Review of Application Progress of Panoramic Imagery in Urban Research

HOU Xin, WANG Yan, WANG Xuan, FAN Wei   

  1. School of Architecture, Tianjin University, Tianjin 300072, China
  • Online:2024-07-01 Published:2024-06-28

摘要: 全景成像技术的进步,街景图像工具的普及,以及人工智能领域的计算机视觉、机器学习和深度学习技术的快速发展,推动了在城市研究中利用全景影像进行大规模、自动化的判别与解析。上述领域的快速发展促使近20年来全景影像、人工智能和城市研究领域之间涌现了大量交叉成果。借助文献计量工具中常用的CiteSpace和VOSviewer作为分析平台,梳理了全景影像在城市研究中的应用进展。首先利用文献共被引聚类网络与术语时区图,划分了全景影像在城市研究中的三个发展阶段。然后借助合著网络和关键词聚类分析,梳理了各阶段全景影像在城市研究中的合著关系、全景影像的获取方式、图像信息的提取技术,归纳了全景影像在城市研究中的四个主要应用领域:城市建成环境、城市景观环境、城市物理环境和智慧城市。最后在历史分期视域下,剖析了促成全景影像应用领域发展的主要驱动因素,并总结了应用全景影像的城市研究目前存在的挑战和未来的发展趋势。

关键词: 全景影像, 街景图像, 城市研究, 人工智能, 深度学习

Abstract: The advances of panoramic imaging technology, the growing popularity of street view imagery tools, and the development in artificial intelligence such as computer vision, machine learning and deep learning, have led to the use of panoramic imagery for large-scale, automated discrimination and parsing in urban research. The rapid development of the aforementioned fields has led to a large number of cross-cutting results between the fields of panoramic imagery, artificial intelligence and urban research in the last two decades. In this paper, CiteSpace and VOSviewer are utilized as analysis platforms to investigate the advancement of panoramic imagery applications in urban research. Firstly, three developmental stages of panoramic imagery in urban research are identified using a literature co-citation clustering network and a terminology time zone map. Subsequently, co-authorship network and keyword clustering analysis are employed to investigate the co-authorship relationships, acquisition methods and image information extraction techniques. Four main application fields of panoramic imagery in urban research are summarized: urban built environment, urban landscape environment, urban physical environment and smart city. Finally, the primary driving factors for the advancement of panoramic imagery application areas are summarized from a historical perspective. Additionally, current challenges and future trends of panoramic imagery application in urban research are discussed.

Key words: panoramic imagery, street view imagery, urban research, artificial intelligence, deep learning