A new realization of adaptive weighted median filters using counter propagation networks

For impulsive noise reduction of an image without the degradation of an original signal, a novel realization of adaptive weighted median filters is developed. The weight controller using counter propagation networks decides an optimal weight of the proposed filter. This controller classifies an input vector into some cluster according to its feature and gives the weights corresponding to the cluster. The parameters in the proposed weight controller are adjustable by using the proposed learning algorithm. The degradation of the original image can be reduced by the proposed technique