A review of multi-temporal remote sensing data change detection algorithms

Jianya Gong, Haigang Sui, Guorui Ma, Qiming Zhou

Research output: Contribution to journalConference articlepeer-review

148 Citations (Scopus)


Change information of the earth's surface is becoming more and more important in monitoring the local, regional and global resources and environment. The large collection of past and present remote sensing imagery makes it possible to analyze spatiotemporal pattern of environmental elements and impact of human activities in past decades. Research has been widely reported on methodology of remote sensing change detection and monitoring. The present reviews have assorted the detection approaches and drawn many useful conclusions. Based on the former classification methods, this article classifies change detection methods from its essence into seven groups, including direct comparison, classification, object-oriented method, model method, time-series analysis, visual analysis and hybrid methods. At the same time, in pre-processing, the effect and methods of geometric correction and radiometric correction is discussed, and in accuracy assessment, this paper summarize the present methods into exterior check and interior check and emphasis on that how to get the ground truth. The challenges that the change detection is currently facing and possible counter measures are also discussed.

Original languageEnglish
Pages (from-to)757-762
Number of pages6
JournalInternational Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives
Publication statusPublished - 2008
Event21st Congress of the International Society for Photogrammetry and Remote Sensing, ISPRS 2008 - Beijing, China
Duration: 3 Jul 200811 Jul 2008

Scopus Subject Areas

  • Information Systems
  • Geography, Planning and Development

User-Defined Keywords

  • Accuracy assessment
  • Change detection
  • Classification framework
  • Remote sensing


Dive into the research topics of 'A review of multi-temporal remote sensing data change detection algorithms'. Together they form a unique fingerprint.

Cite this