Removal of Salt and Pepper Noise in Images using Modified Directional Filter
Thresa Paulraj · 2013
In this paper, an algorithm based on directional filtering for suppressing salt and pepper noise is proposed. This algorithm is similar to the modified directional weighted median filter proposed by Ching-Ta-Lu et al but has been modified to achieve better subjective quality images with better edge preservation capability ,higher peak signal to noise ratio(PSNR) and lower execution time. The algorithm of Ching et al uses directional weighted median filter whereas the proposed algorithm uses directional nearest neighbour pixel identification method for filtering. The results obtained using the proposed algorithm are compared with those obtained by Ching et al in terms of subjective quality, PSNR and execution time for 80% noise density. It is found that the proposed algorithm yields better results and also preserves the edge details to a greater extent.