The translation sensitivity of wavelet-based registration

Author

Stone, Harold S. ; Le Moigne, Jacqueline ; McGuire, Morgan

Author_Institution

NEC Res. Inst., Princeton, NJ, USA

Volume

21

Issue

10

fYear

1999

fDate

10/1/1999 12:00:00 AM

Firstpage

1074

Lastpage

1081

Abstract

This paper studies the effects of image translation on wavelet-based image registration. The main result is that the normalized correlation coefficients of low-pass Haar and Daubechies wavelet subbands are essentially insensitive to translations for features larger than twice the wavelet blocksize. The third-level low-pass subbands produce a correlation peak that varies with translation from 0.7 and 1.0 with an average in excess of 0.9. Translation sensitivity is limited to the high-pass subband and even this subband is potentially useful. The correlation peak for high-pass subbands derived from first and second-level low-pass subbands ranges from about 0.0 to 1.0 with an average of about 0.5 for Daubechies and 0.7 for Haar. We use a mathematical model to develop these results, and confirm them on real data

Keywords

Haar transforms; correlation theory; image registration; wavelet transforms; correlation peak; high-pass subband; image translation; low-pass Daubechies wavelet subbands; low-pass Haar wavelet subbands; normalized correlation coefficients; translation sensitivity; wavelet blocksize; wavelet-based image registration; Biomedical imaging; Costs; Image registration; Image resolution; Low pass filters; Mathematical model; Pathology; Pixel; Satellite broadcasting; Wavelet transforms;

fLanguage

English

Journal_Title

Pattern Analysis and Machine Intelligence, IEEE Transactions on

Publisher

ieee

ISSN

0162-8828

Type

jour

DOI

10.1109/34.799911

Filename

799911