We present a novel method for irregular pyramid construction. The cont
ributions of this paper are twofold: (1) Instead of starting with the
original. regular pixel grid, we first apply an adaptive Voronoi tesse
llation to the image. This provides the advantage that the number of c
ells at the bottom level of the pyramid is already reduced as compared
to the number of pixels of the original image, (2) For the constructi
on of the irregular pyramid, we present a Hopfield neural network whic
h controls the decimation process. The decimation by Hopfield networks
includes naturally the concept of the adaptive pyramid. The validity
of our approach is demonstrated by several examples in image segmentat
ion. (C) 1996 Academic Press, Inc.