This paper proposes a new method using an artificial neural network to remove random-valued impulse noise (RVIN) in images. The inputs of the neural model used to detect the RVIN are formed using basic and related gradient values. The detection of the noisy pixels is realized in 3 phases using the proposed neural detector. In order to obtain a more robust detector, 2 different networks, which are trained with an artificial training image corrupted with high and low clutter densities, are used. The extensive simulation results show that the proposed method is significantly better than the compared filters in terms of its image restoration and noise detection performance.
: This paper proposes a new method using an arti?cial neural network to remove random-valued impulse noise
(RVIN) in images. The inputs of the neural model used to detect the RVIN are formed using basic and related gradient
values. The detection of the noisy pixels is realized in 3 phases using the proposed neural detector. In order to obtain a
more robust detector, 2 di?erent networks, which are trained with an arti?cial training image corrupted with high and
low clutter densities, are used. The extensive simulation results show that the proposed method is signi?cantly better
than the compared ?lters in terms of its image restoration and noise detection performance.
