Change detection of urban objects using 3D point clouds: A review

Uwe Stilla, Yusheng Xu

Research output: Contribution to journalReview articlepeer-review

53 Scopus citations

Abstract

Over recent decades, 3D point clouds have been a popular data source applied in automatic change detection in a wide variety of applications. Compared with 2D images, using 3D point clouds for change detection can provide an alternative solution offering different modalities and enabling a highly detailed 3D geometric and attribute analysis. This article provides a comprehensive review of point-cloud-based 3D change detection for urban objects. Specifically, in this study, we had two primary aims: (i) to ascertain the critical techniques in change detection, as well as their strengths and weaknesses, including data registration, variance estimation, and change analysis; (ii) to contextualize the up-to-date uses of point clouds in change detection and to explore representative applications of land cover and land use monitoring, vegetation surveys, construction automation, building and indoor investigations, and traffic and transportation monitoring. A workflow following the PRISMA 2020 rules was applied for the search and selection of reviewed articles, with a brief statistical analysis of the selected articles. Additionally, we examined the limitations of current change detection technology and discussed current research gaps between state-of-the-art techniques and engineering demands. Several remaining issues, such as the reliability of datasets, uncertainty in results, and contribution of semantics in change detection, have been identified and discussed. Ultimately, this review sheds light on prospective research directions to meet the urgent needs of anticipated applications.

Original languageEnglish
Pages (from-to)228-255
Number of pages28
JournalISPRS Journal of Photogrammetry and Remote Sensing
Volume197
DOIs
StatePublished - Mar 2023

Keywords

  • Applications
  • Change detection
  • Point clouds
  • Urban objects

Fingerprint

Dive into the research topics of 'Change detection of urban objects using 3D point clouds: A review'. Together they form a unique fingerprint.

Cite this