Remote sensing image registration algorithm based on SIFT

Chuanbin Deng, Lei Guo, Wei Li

Research output: Contribution to journalArticlepeer-review

5 Scopus citations

Abstract

To resolve multi-sensor remote sensing image registration,an method based on scale invariant features transformCSIFT)is proposed. First,the local invariant features of images are extracted,Ratio of the first and the second nearest neighbor distance with mutual correspondence constraint is used to setup the initial correspondences. Then,Random Sample Concensus algorithm is used to remove the mismatched feature points. Experiments demonstrate automatic registration of multi-sensor remote sensing images and images with different resolutions can be achived.

Original languageEnglish
Pages (from-to)1742-1747
Number of pages6
JournalChinese Journal of Sensors and Actuators
Volume22
Issue number12
StatePublished - Dec 2009

Keywords

  • Image registration
  • Mutual correspond dence constraint
  • Random sample concensus
  • Scale invariant features transform

Fingerprint

Dive into the research topics of 'Remote sensing image registration algorithm based on SIFT'. Together they form a unique fingerprint.

Cite this