In the last decade, multiscale techniques for gray-level texture analysis h
ave been intensively used. In this paper, we aim to extend these techniques
to color images. We introduce wavelet energy-correlation signatures and we
derive the transformation of these signatures upon linear color space tran
sformations. Experiments are conducted on a set of 30 natural colored textu
re images in which color and gray-level texture classification performances
are compared. It is demonstrated that the wavelet correlation features con
tain more information than the intensity or the energy features of each col
or plane separately. The influence of image representation in color space i
s evaluated. (C) 1999 Pattern Recognition Society. Published by Elsevier Sc
ience Ltd. All rights reserved.