Recently local frequency has been suggested as an efficient feature for performing image analysis and multi-modal image registration. While local frequency representation can detect the structure of the scene in the image (ridges and edges simultaneously), its computation typically requires using a bank of analytic filters such as two-dimensional (2-D) Gabor filters characterized by a set of parameters (spatial frequency, orientation angle and the filter bandwidth). Tuning a set of Gabor filters that can cover the entire frequency space for the given image is critical for obtaining an accurate representation of the local frequency; however, choosing the size of the filter bank is often done in an ad hoc manner and one usually errs on the conservative side by employing a larger than needed set of filters in order to cover the entire range of frequencies, thus sacrificing computational efficiency. In this paper, we propose a computationally efficient approach for developing the local frequency representation from given images. The new algorithm employs a filter bank of size four, whose parameters are adaptively selected based on the image content. Motivation for the new scheme comes from biological considerations and it is shown that an accurate local frequency representation can be developed from a fusion of the four Gabor filter outputs. Among the notable advantages of the proposed algorithm are reduction in computational requirements (which makes it ideal for the registration and fusion of large-format images) and its robustness to specific scene details.
[1]
Dennis Gabor,et al.
Theory of communication
,
1946
.
[2]
Boualem Boashash,et al.
Estimating and interpreting the instantaneous frequency of a signal. II. A/lgorithms and applications
,
1992,
Proc. IEEE.
[3]
Hans Knutsson,et al.
Signal processing for computer vision
,
1994
.
[4]
J. Daugman.
Spatial visual channels in the fourier plane
,
1984,
Vision Research.
[5]
Anil K. Jain,et al.
Registering Landsat images by point matching
,
1989
.
[6]
Boualem Boashash,et al.
Estimating and interpreting the instantaneous frequency of a signal. I. Fundamentals
,
1992,
Proc. IEEE.
[7]
Jonathan Liu,et al.
Multimodal image registration using local frequency
,
2000,
Proceedings Fifth IEEE Workshop on Applications of Computer Vision.
[8]
Eric Mjolsness,et al.
New Algorithms for 2D and 3D Point Matching: Pose Estimation and Correspondence
,
1998,
NIPS.
[9]
Jundong Liu,et al.
Local frequency representations for robust multimodal image registration
,
2002,
IEEE Transactions on Medical Imaging.
[10]
Xosé R. Fernández-Vidal,et al.
The Selection of Natural Scales in 2D Images Using Adaptive Gabor Filtering
,
1998,
IEEE Trans. Pattern Anal. Mach. Intell..
[11]
N. Graham.
Visual Pattern Analyzers
,
1989
.