An architecture of a neural network with assembly organization is desc
ribed. Such network architecture is applied to the problem of texture
segmentation bn natural scenes. The network is partitioned into severa
l sub-networks. Each subnetwork is a column structure in which feature
s me represented by means of ''float'' coding. Input data excite corre
sponding ''floats'' of neurons in the subnetworks. In the process of l
earning the weights of modifiable connections between excited neurons
are changed so that Hebb's assemblies are formed in the column structu
res. All subnetworks me incorporated into a single network by a neural
activity control system. Computer simulation of the proposed network
has been performed. The results of computer simulations show the possi
bility of successful application of the assembly neural network to the
problem of texture segmentation. Copyright (C) 1996 Elsevier Science
Ltd