PCNN Based Hybrid Approach for Suppression of High Density of Impulsive Noise

Abstract

Many image processing applications requires Impulsive noise elimination. Windyga’s peak-and-valley filter used to remove impulsive noise, its main disadvantage is that it works only for low density of noises. In this Paper, a variation of the two dimensional peak-and-valley filters is proposed to overcome this problem. It is based on minimum/maximum values present in the noisy image, which replaces the noisy pixel with a value based on neighborhood information based on the outcomes of PCNN (Pulse Coupled Neural Network). This method preserves constant and edge areas even under high impulsive noise probability. Extensive Computer simulations show that the proposed approach outperforms other filters in the noise reduction and the image details preservation.