Skip to main navigation Skip to search Skip to main content

WT-based rotation compensated image correlation matching method with sub-pixel resolution

  • Jingying Jiang*
  • , Qilian Yu
  • , Kexin Xu
  • , Xiaodong Hu
  • *Corresponding author for this work
  • Tianjin University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

The algorithm proposed is based upon phase correlation with rotational compensation and paraboloid surface fit and can be used to determine the translational and rotational parameters needed to match the two images. The approximate signals obtained through the wavelet transform the images are used to find the rotational angle between the two images, and phase correlation is then performed; finally, the paraboloid surface fit method is used to find the sub-pixel translation between them. The proposed algorithm has been experimentally and analytically shown, to be accurate to more than 1/50-pixel. The algorithm presented has been applied to match palm print images with validity and accuracy demonstrated.

Original languageEnglish
Title of host publicationProceedings of the Second International Symposium on Instrumentation Science and Technology
EditorsT. Jiubin, W. Xianfang, T. Jiubin, W. Xianfang
Pages3/527-3/531
StatePublished - 2002
Externally publishedYes
EventProceedings of the second International Symposium on Instrumentation Science and Technology - Jinan, China
Duration: 18 Aug 2002 → 22 Aug 2002

Publication series

NameProceedings of the Second International Symposium on Instrumentation Science and Technology
Volume3

Conference

ConferenceProceedings of the second International Symposium on Instrumentation Science and Technology
Country/TerritoryChina
CityJinan
Period18/08/02 → 22/08/02

Keywords

  • Image correlation matching
  • Phase correlation
  • Sub-pixel resolution
  • Wavelet transform (WT)

Fingerprint

Dive into the research topics of 'WT-based rotation compensated image correlation matching method with sub-pixel resolution'. Together they form a unique fingerprint.

Cite this