This paper presents a new rotation-invariant image retrieval method,
which extends a recently introduced classification technique based
on steerable wavelet transforms. In the proposed procedure, the feature
extraction step consists of estimating the covariations (lower-order
cross-correlations) between the wavelet subband coefficients, which
are modeled as subGaussian random vectors. The similarity measurement
is carried out first by employing norms calculating the distance
between the covariation matrices representing two distinct images
and second by evaluating the Kullback-Leibler Distance (KLD) between
their corresponding subGaussian distributions. We provide analytical
expressions relating the subGaussian features corresponding to a
rotated image from the features of the original image. Finally, we
relate the employed optimal lower-order correlation (p/spl les/2)
to the degree of nonGaussianity of the wavelet coefficients, and
we demonstrate the effectiveness of our method using real texture
images.
%0 Conference Paper
%1 Tzagkarakis2004
%A Tzagkarakis, G.
%A Beferull-Lozano, B.
%A Tsakalides, P.
%B Signals, Systems and Computers, 2004. Conference Record of the Thirty-Eighth
Asilomar Conference on
%D 2004
%I IEEE Computer Society
%K Gaussian KLD, Kullback-Leibler analytical classification classification, coefficient correlation correlation, covariance covariation distance, distribution, estimation, expression, extraction, feature image image, images, invariant lower-order matrices matrices, measurement, optimal random real realistic retrieval, rotation, similarity steerable subGaussian subband technique, texture texture, theory, transform, transforms vector, wavelet
%P 397--401
%R 10.1109/ACSSC.2004.1399161
%T SubGaussian rotation-invariant features for steerable wavelet-based
image retrieval
%V 1
%X This paper presents a new rotation-invariant image retrieval method,
which extends a recently introduced classification technique based
on steerable wavelet transforms. In the proposed procedure, the feature
extraction step consists of estimating the covariations (lower-order
cross-correlations) between the wavelet subband coefficients, which
are modeled as subGaussian random vectors. The similarity measurement
is carried out first by employing norms calculating the distance
between the covariation matrices representing two distinct images
and second by evaluating the Kullback-Leibler Distance (KLD) between
their corresponding subGaussian distributions. We provide analytical
expressions relating the subGaussian features corresponding to a
rotated image from the features of the original image. Finally, we
relate the employed optimal lower-order correlation (p/spl les/2)
to the degree of nonGaussianity of the wavelet coefficients, and
we demonstrate the effectiveness of our method using real texture
images.
@inproceedings{Tzagkarakis2004,
abstract = {This paper presents a new rotation-invariant image retrieval method,
which extends a recently introduced classification technique based
on steerable wavelet transforms. In the proposed procedure, the feature
extraction step consists of estimating the covariations (lower-order
cross-correlations) between the wavelet subband coefficients, which
are modeled as subGaussian random vectors. The similarity measurement
is carried out first by employing norms calculating the distance
between the covariation matrices representing two distinct images
and second by evaluating the Kullback-Leibler Distance (KLD) between
their corresponding subGaussian distributions. We provide analytical
expressions relating the subGaussian features corresponding to a
rotated image from the features of the original image. Finally, we
relate the employed optimal lower-order correlation (p/spl les/2)
to the degree of nonGaussianity of the wavelet coefficients, and
we demonstrate the effectiveness of our method using real texture
images.},
added-at = {2011-03-27T19:35:34.000+0200},
author = {Tzagkarakis, G. and Beferull-Lozano, B. and Tsakalides, P.},
biburl = {https://www.bibsonomy.org/bibtex/283c5c69f6297b89ca7f5a88f019c1468/cocus},
bookitleaddon = {Novemeber 7--10, 2004},
booktitle = {Signals, Systems and Computers, 2004. Conference Record of the Thirty-Eighth
Asilomar Conference on},
doi = {10.1109/ACSSC.2004.1399161},
file = {:./01399161.pdf:PDF},
interhash = {a2ebd4b2eea24ee054db4b926358b8de},
intrahash = {83c5c69f6297b89ca7f5a88f019c1468},
keywords = {Gaussian KLD, Kullback-Leibler analytical classification classification, coefficient correlation correlation, covariance covariation distance, distribution, estimation, expression, extraction, feature image image, images, invariant lower-order matrices matrices, measurement, optimal random real realistic retrieval, rotation, similarity steerable subGaussian subband technique, texture texture, theory, transform, transforms vector, wavelet},
location = {#ieeeaddr#},
pages = { 397--401},
publisher = {{IEEE} Computer Society},
timestamp = {2011-03-27T19:35:44.000+0200},
title = {SubGaussian rotation-invariant features for steerable wavelet-based
image retrieval},
venue = {Pacific Grove, CA, USA},
volume = 1,
year = 2004
}