M. Muneyasu et al., IMAGE-RESTORATION USING LAYERED NEURAL-NETWORKS AND HOPFIELD NETWORKS, Electronics and communications in Japan. Part 3, Fundamental electronic science, 78(5), 1995, pp. 77-88
An algorithm is developed for the restoration of an image degraded by
a known two-dimensional (2-D) shift-invariant point-spread function an
d corrupted with white noise. A layered neural network and the Hopfiel
d network are used for the edge detection, and the restoration and smo
othing of a blurred image, respectively. In particular, a layered neur
al network is proposed for exact edge detection where the inputs consi
st of three pixel values and a local variance in a 2-D mask. This netw
ork can detect edges and suppress the noise in an image at the same ti
me, and its performance is adjusted by learning. Finally, some example
s are given to illustrate the utility of the proposed algorithm.