The Translation Sensitivity of Wavelet-Based Registration

Authors:
Harold S. Stone;Jacqueline Le Moigne;Morgan McGuire
Affiliations:
NEC Research Institute, Princeton, NJ;NASA Goddard Space Flight Center, Greenbelt, MD;Massachusetts Institute of Technology, Cambridge
Venue:
IEEE Transactions on Pattern Analysis and Machine Intelligence
Year:
1999

Citing 7
Cited 2

A survey of image registration techniques

ACM Computing Surveys (CSUR)
Multiresolution image registration

ICIP '95 Proceedings of the 1995 International Conference on Image Processing (Vol. 3)-Volume 3 - Volume 3
Image registration using both distance and angle information

ICIP '95 Proceedings of the 1995 International Conference on Image Processing (Vol. 3)-Volume 3 - Volume 3
Shift invariant wavelet packet bases

ICASSP '95 Proceedings of the Acoustics, Speech, and Signal Processing, 1995. on International Conference - Volume 02
Progressive wavelet correlation using Fourier methods

IEEE Transactions on Signal Processing
Template matching based on a grayscale hit-or-miss transform

IEEE Transactions on Image Processing
An FFT-based technique for translation, rotation, and scale-invariant image registration

IEEE Transactions on Image Processing

Rotation-invariant pattern matching using wavelet decomposition

Pattern Recognition Letters
On the shiftability of dual-tree complex a globally optimal bilinear programming approach to the design of approximate hilbert pairs of orthonormal wavelet bases

IEEE Transactions on Signal Processing

Quantified Score

Hi-index	0.15

Visualization

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.